ArticlePDF Available

Machine Learning Adoption in Blockchain-Based Smart Applications: The Challenges, and a Way Forward

Authors:
  • ICT-University

Abstract and Figures

In recent years, the emergence of blockchain technology (BT) has become a unique, most disruptive, and trending technology. The decentralized database in BT emphasizes data security and privacy. Also, the consensus mechanism in it makes sure that data is secured and legitimate. Still, it raises new security issues such as majority attack and double-spending. To handle the aforementioned issues, data analytics is required on blockchain based secure data. Analytics on these data raises the importance of arisen technology Machine Learning (ML). ML involves the rational amount of data to make precise decisions. Data reliability and its sharing are very crucial in ML to improve the accuracy of results. The combination of these two technologies (ML and BT) can provide highly precise results. In this paper, we present a detailed study on ML adoption for making BT-based smart applications more resilient against attacks. There are various traditional ML techniques, for instance, Support Vector Machines (SVM), clustering, bagging, and Deep Learning (DL) algorithms such as Convolutional Neural Network (CNN) and Long short-term memory (LSTM) can be used to analyse the attacks on a blockchain-based network. Further, we include how both the technologies can be applied in several smart applications such as Unmanned Aerial Vehicle (UAV), Smart Grid (SG), healthcare, and smart cities. Then, future research issues and challenges are explored. At last, a case study is presented with a conclusion. INDEX TERMS Blockchain, machine learning, smart grid, data security and privacy, data analytics, smart applications.
Content may be subject to copyright.
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.DOI
Machine Learning adoption in
Blockchain-based Smart Applications:
The challenges, and a way forward
SUDEEP TANWAR1, QASIM BHATIA1, PRUTHVI PATEL1, APARNA KUMARI1, PRADEEP
KUMAR SINGH2, WEI-CHIANG HONG3, SENIOR MEMBER, IEEE
1Department of Computer Science and Engineering, Institute of Technology, Nirma University, Ahmedabad, Gujarat, India (e-mails:
sudeep.tanwar@nirmauni.ac.in, 16bce059@nirmauni.ac.in, 16bce055@nirmauni.ac.in, 17ftphde22@nirmauni.ac.in)
2Department of Computer Science & Engineering, Jaypee University of Information Technology, Waknaghat, Solan (H.P.) India (e-mail:
pradeep_84cs@yahoo.com)
3Department of Information Management, Oriental Institute of Technology, Panchiao, New Taipei, Taiwan. (email: samuelsonhong@gmail.com)
Corresponding author: Wei-Chiang Hong (e-mail: samuelsonhong@gmail.com).
ABSTRACT In recent years, the emergence of blockchain technology (BT) has become a unique, most
disruptive, and trending technology. The decentralized database in BT emphasizes data security and privacy.
Also, the consensus mechanism in it makes sure that data is secured and legitimate. Still, it raises new
security issues such as majority attack and double-spending. To handle the aforementioned issues, data
analytics is required on blockchain based secure data. Analytics on these data raises the importance of arisen
technology Machine Learning (ML). ML involves the rational amount of data to make precise decisions.
Data reliability and its sharing are very crucial in ML to improve the accuracy of results. The combination
of these two technologies (ML and BT) can provide highly precise results. In this paper, we present a
detailed study on ML adoption for making BT-based smart applications more resilient against attacks.
There are various traditional ML techniques, for instance, Support Vector Machines (SVM), clustering,
bagging, and Deep Learning (DL) algorithms such as Convolutional Neural Network (CNN) and Long
short-term memory (LSTM) can be used to analyses the attacks on a blockchain-based network. Further, we
include how both the technologies can be combined in several smart applications such as Unmanned Aerial
Vehicle (UAV), Smart Grid (SG), healthcare, and smart cities. Then, future research issues and challenges
are explored. At last, a case study is presented with a detailed conclusion.
INDEX TERMS Blockchain, Machine Learning, Smart Grid, Data Security and Privacy, Data Analytics,
Smart Applications
I. INTRODUCTION
FROM the past few decades, data has become an essential
source of intelligence and carries new opportunities
to the real-life problems such as wireless communications,
bioinformatics [1], agriculture [2], and finance [3] through
smart applications. These applications are data-driven and
incorporate actionable insights into user experience, which
enables individuals to complete the desired task more effi-
ciently. It operationalizes insights, personalizes the customer
experience, optimizes customer interactions, improves op-
erational efficiency, and enable new business model. There
are various smart applications such as SG, UAV, Smart
Cities, which makes the life of an individual easier. These
applications generate a huge amount of data, and storage
of this ever-evolving data in databases is a problem, and
its communication also raises security concerns. To handle
these issues, BT can be used, which has a distributed database
network. It was coined by Satoshi Nakamoto in the year 2008
and contained a time-stamped series of tamper-proof records,
which are managed by a cluster of distributed computers.
It comprises of a chain of blocks that are connected using
cryptographic primitives. The three mainstays of BT are
immutability, decentralization, and transparency. These three
characteristics opened its door for a wide variety of applica-
tions, for example, digital currency existence (currency with
no physical existence) and analysis on the suitability of it
in smart applications [4]. Although BT ensures security and
privacy issues, some vulnerabilities also started appearing
after its implementation. For instance, the nature of attacks
began to be increasingly sophisticated such as majority
VOLUME 4, 2016 1
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
attacks (51% attack) that control voting, Sybil attacks for fake
identity generation to control the consensus [5]. To handle the
aforementioned issue, a robust Intrusion Detection System
(IDS) is required in place because the traditional methods use
a signature-based approach to detect specific patterns. But,
to detect intrusions and attack patterns, one of the emerging
technology known as ML can be used to analyze the data
traffic. Thus, designing efficient and effective algorithms
to analyze this massive amount of data is in dire need of
handling the blockchain-based smart applications. Hence,
ML is highly prevalent today and uses a dozen times a
day without even realizing it. ML encompasses computers
to study, think, and act without intervention of humans.
It is considered to be one of the applications of Artificial
Intelligence (AI). ML provides computers the competency to
learn without being programmed it explicitly. Its basic idea
is to build an efficient algorithm that can accept input data
and, with the help of statistical analysis, make a prediction,
and update the outputs. A substantial amount of data can be
analyzed by ML to create data-driven decisions.
In a communication network of blockchain-based smart
applications, there is layer-wise handling of security issues.
Some security issues are handled at the network layer, such as
malicious packets and some at the application layer such as
malware [6]. At the network layer, malicious packets can be
used to impose the network to establish fraudulent consensus.
A naive solution to this problem can be to use a firewall
to ensure that packets meet pre-defined security criteria [7].
Though, the attacks are becoming more sophisticated with
unseen patterns to bypass a firewall. To prevent this issue,
packets header data can be analyzed using ML models [8] in
real-time using historical data. This analysis helps to detect
new and changing patterns. Similarly, ML techniques can
be used to classify malware to end-point such as servers,
mobile, or workstations. Further, several blockchain-based
smart applications such as UAV [9], Data Trading [10], SG
build trust between data exchangers [11]. It is very crucial
in any smart application at the same time; data should be
secure. BT ensures data security but to build confidence,
and ML techniques are used to predict untrustworthy nodes
based on past patterns. Similarly, UAVs have significantly
different network topology compared to the conventional
blockchain network topology [9]. It includes communication
using satellites and various ground stations. For UAV, BT is
used to securely store coordinates and other relevant data to
maintaining graph integrity for the vehicles. In subsequent
sections, we explore the recent research work on ML adop-
tion in the blockchain-based smart application.
A. SCOPE OF THIS SURVEY
Several survey work has been published till date by the
different researchers on the various aspects of the adoption
of ML in blockchain-based smart applications [5]. As per
our knowledge, most of the surveys have focused on specific
areas, fields, or applications where it requires both ML and
blockchain. The proposed survey covers all the fundamental
aspects of ML to be applied BT based applications, for
instance, intrusion detection. A review conducted by Meng
et al. [12] describes how blockchain help to meet intrusion
detection. The authors mitigate trust issues by establishing a
collaborative IDS using smart contracts [17]. Further, Conti
et al. [13] surveyed on security and privacy issues of bitcoin.
They discuss various categories of attacks, such as double-
spending attacks, client-side security threats, and mining
pool attacks. To handle the aforementioned issue, Rahouti et
al. [5] discuss specific ML-based solutions and discuss the
handling of specific social issues such as human trafficking
and drug sales through cryptocurrency. Then, Ucci et al [14]
explored malware analysis using ML techniques. Features of
malware were thoroughly discussed, and a detailed taxonomy
has been proposed. Salah et al. [15] and Casino et al. [16]
conducted a review on blockchain-based applications. To
clarify the main difference between other surveys paper and
this survey paper, a comprehensive comparison has been
shown in Table 1. It includes objectives, merits, and demerits
of peer surveys concerning numerous parameters such as
architecture, applications, open issues, taxonomy, and merits,
demerits of the existing approaches. A master taxonomy of
ML for BT is summarized in Figure 3.
B. RESEARCH CONTRIBUTIONS
Though several research works exist to address the various
ML usages for the blockchain-based system but It has not
been exploited to its full potential. In this paper, we investi-
gated the ML usages for BT. The following are the research
contribution of this paper.
A brief discussion on how ML and blockchain can be
used together with a proposed architecture diagram.
To develop taxonomy covering ML techniques required
for BT based environment. In each part of the taxonomy,
existing work has been discussed in detailed to handle
several issues, such as preventing and predicting attacks
on the blockchain network.
A case study is presented to demonstrate the ML tech-
niques for blockchain-based smart applications, for ex-
ample, SG.
C. ORGANIZATION
The rest of the paper organized as follows: Section 2 provides
a background of Blockchain, ML, and list all benefits while
applying ML for BT based application. Then, we proposed an
architecture. Survey procedures and taxonomy presented in
Section 3. Then, in Section 4-7, we present existing advance-
ments in ML approaches for BT-based smart applications.
A discussion on future research issues and challenges is
presented in Section 8. Further, section 9 covers a case study
on the SG system, and finally, we concluded the paper.
II. BACKGROUND
2VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
TABLE 1: Comparison of the existing surveys with the proposed survey.
Authors Year Objectives of Survey Merits Demerits 1 2 3 4 5
Meng et al. [12] 2018 To present use of blockchain in in-
trusion detection
Scope of application of blockchain
was discussed
Discusses only data sharing and
trust management issues of collab-
orative intrusion detection
XXXXX
Conti et al. [13] 2018 To discuss various security and pri-
vacy issues in bitcoin
A Comprehensive review of possi-
ble attacks on bitcoin and provided
countermeasures
Blockchain issues are not high-
lighted
XXXXX
Rahouti et al. [5] 2018 Survey on ML security solutions for
bitcoin
In-depth and wide classification of
major threats and extensive expla-
nation of the role of ML
Other applications of blockchain
are missing
XXXXX
Ucci et al. [14] 2018 To study ML techniques for mal-
ware analysis
Time and space complexity for var-
ious methodologies has been de-
scribed in detail
Lacks discussion on uses of these
techniques in a blockchain environ-
ment
XXXXX
Salah et al. [15] 2019 Discuss applications, platforms,
and protocols in blockchain
specifically for AI
The decentralization feature of
blockchain is explained with a spe-
cific view of AI
Discussion on privacy is not cov-
ered in detail
XXXXX
Casino et al. [16] 2019 Review blockchain-based applica-
tions and identify open issues
Prerequisites for blockchain appli-
cations are thoroughly discussed
Focused on applications, not the
open issues
XXXXX
Proposed Survey - To survey how ML can be used
in blockchain-based smart applica-
tions
Discusses architecture and technol-
ogy at a fundamental level and
bridges the gap between two tech-
nology
-XXXXX
Parameters- 1:Architecture, 2:Application, 3:Open Issues and Challenges, 4:Taxonomy, 5:Security
Notations- X: considered, and X: not considered.
A. MACHINE LEARNING
ML is the field of study that focuses on building applications
that learn through experience. It is the ability to teach a
computer without programming it explicitly [18]. ML en-
compasses its work from a diverse set of disciplines, includ-
ing philosophy, information theory, probability and statistics,
control theory, psychology and neurobiology, computational
complexity, and artificial intelligence [19]. ML algorithms
are used in many application and benefited it as listed below:
In Data mining, large databases contain different pat-
terns that can be discovered automatically by using ML
techniques to analyze outcomes, for instance, medical
treatments of a patient from health record databases
or to identify the creditworthiness of a person from
financial databases.
ML applies in areas where a deterministic algorithm
is not promising, such as human face recognition from
images.
Application domains where the adaptable programming
is required, for instance, controlling manufacturing pro-
cesses as per the demand of the customer and adapting
to the varying reading interests of readers.
ML algorithm eare application specific and depends on the
output required by the system. There are several ML algo-
rithms, such as Supervised ML, Semi-Supervised ML, and
Unsupervised ML. (i) Supervised ML uses statistical models
to predict output in numerical data and classify the correct la-
bel [20]. Here, the most commonly known algorithms include
the regression approach and decision trees. (ii) Unsupervised
ML does not have label data. Here, data samples are grouped
into clusters depending on their similarity or dissimilarity
[21] using a different approach. For example, K means clus-
tering and association rules algorithms. (iii) Semi-Supervised
ML is also a category of interest [22]. It involves a mixture of
supervised and unsupervised ML techniques. Unsupervised
learning may be applied to discover the structure of input
variables, following which it is used to make best guess
predictions for the unlabeled data. It feeds that predicted data
back into the supervised ML algorithm as training data and
use the model to make predictions on unseen data.
B. BLOCKCHAIN
Blockchains are an immutable set of records that are cryp-
tographically linked together for audit [23]. It is similar to
an accounting ledger. Here, previous records in accounting
ledger cannot be changed, and new records need to be veri-
fied by a trusted party. The only difference between these two
is that new blocks (set of records) checked by a decentralized
structure of nodes that have a copy of the ledger. There
is no centralized party to verify the records. Blockchain is
formed by linking valid blocks together; the current block
contains the hash of the previous block, and so on, as shown
in FIGURE 1. This makes blockchain traceable and resistant
to change [24]. Older blocks cannot be modified, in case
they are changed in any way; their hash would change. This
emphasis to link hash in all subsequent blocks to make the
blockchain network valid again. A copy of the blockchain is
available with every individual within the network; hence-
forth, any changes can be cross verified by the other users.
These copies of the blockchain are updated with the addition
of a new block. Then, everyone can see the block, depending
on the permissions assigned by the administrator. BT uses
a cryptographic secure hash algorithm (SHA) such as SHA-
256 and SHA-512 to maintain the data integrity within the
block. Each block has a unique hash value. For instance,
Ethereum uses Keccak-256 and Keccak-512, while Bitcoin
uses double SHA-256. This SHA is a collision-resistant al-
gorithm, where no two different input data could produce the
same output (hash value). Henceforward SHA can be used to
check if the data is the same or not. There are various SHA
algorithms developed by the NSA and NIST and belongs
to the SHA-2 family [25]. SHA was initially envisioned
VOLUME 4, 2016 3
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
Block Hash
Difficulty
Previous Block Hash
Block Header
Nonce
Timestamp
Merkle Root
Transaction Hashes
Block Hash
Difficulty
Previous Block Hash
Block Header
Nonce
Timestamp
Merkle Root
Transaction Hashes
Block Hash
Difficulty
Previous Block Hash
Block Header
Nonce
Timestamp
Merkle Root
Transaction Hashes
- - - - - - - - - - - - - -
- - - - - - - - - - - - -
Block-N
Block-2
Block-1
FIGURE 1: Blockchain Structure
as a fragment of the Digital Signature Standard (DSS) to
produce the signature. Further, consensus algorithms are
used to determine block validity. The selection of algorithms
depends on the type of blockchain, such as public, private,
and consortium blockchain. The selected algorithm should
ensure the consensus among nodes [26]. It must be able to
use resources efficiently and tolerate a degree of safety during
the event of attacks.
Further, a smart contract is a program (set of codes) that
runs on the blockchain and adds blocks whenever certain
conditions are met. It is defined by translating actual legal
contracts into programs to enforce the legal contract onto
the blockchain [27]. It is similar to stored-procedures in
relational databases. It is stored as scripts in the blockchain
and executed according to the data fed to them to produce
outputs that are expected from the original contract [28]. It
governs transactions either executed fully or partially based
on the current input. The primary purpose of a smart contract
is to provide superior security with a reduction in cost and
delays associated with traditional contracts.
C. INTEGRATION OF MACHINE LEARNING IN
BLOCKCHAIN-BASED APPLICATIONS
The learning capabilities of ML can be applied to blockchains
based applications to make them smarter. By using ML
security of the distributed ledger may be improved. ML may
also be used to enhance the time taken to reach consensus
by building better data sharing routes. Further, it creates an
opportunity to build better models by taking advantage of the
decentralized architecture of BT. We proposed architecture
for ML adoption in BT-based smart application, as shown in
Figure 2. Here, the smart application collects data from dif-
ferent data sources such as sensors, smart devices, and Intenet
of Things (IoT) devices. Data collected from these devices
get processed as part of smart applications. The blockchain
work as an integral part of these smart applications. Then,
ML can be applied to these application’s data for analysis
(Data analytics and real-time analytics) and prediction. The
data sets used by ML models could be stored on a blockchain
network. This reduces errors in the data such as duplica-
tion, missing data value, errors, and noise. Blockchains are
focused on the data, and hence data-related issues in ML
FIGURE 2: Proposed architecture for Machine Learning
adoption in Blockchain-based Smart applications
models may be eliminated. ML-models can be based on
specific segments of the chain rather than the entire data-
set. This could give custom models for different applications,
such as fraud detection and identity theft detection. Few of
the benefit are listed beneath when ML is applied:
User authentication as a legitimate user for requesting or
performing any transaction in the blockchain network.
BT provides a high level of security and trust.
Blockchain integrates public ML models into smart
contracts to ensure that the conditions and terms which
were previously agreed are sustained.
BT helps in the reliable implementation of an incentive-
based system; thus, it encourages users/customers to
contribute data. This huge data will help to improve ML
model performance.
ML models can be updated on-chain environment of BT
with a small fee and off-chain, locally on an individu-
alâ ˘
A´
Zs device without any costs
Good data contributions can happen from users/customers,
these data consistently computed, and rewards can be
given to the users.
Tamper-proof smart contracts can be evaluated by dif-
ferent machines (having different hardware configura-
tion), ML models will not diverge from their potential
and produce results exactly as it is supposed to do.
Payments processed in real-time with trust on a
blockchain environment.
Blockchains tools, for instance, Ethereum, deal with
thousands of decentralized machines all over the world.
This guaranteed users that it is never completely un-
reachable or offline.
4VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
III. SURVEY PROCEDURES AND TAXONOMY
In this section, we refer to the procedures used to perform this
study; for example, our search approach and inclusion criteria
for the final set of papers. Likewise, we present a detailed
taxonomy based on our literature review.
A. SYSTEMATIC LITERATURE REVIEW
We used standard databases (for example IEEEXplore, ACM
Digital Library, ScienceDirect, and Springerlink) and Google
Scholar to search existing research work, using keywords
such as â ˘
AIJMachine Learning for Blockchain-based Smart
Application,â ˘
A˙
I ˘
AIJBlockchain for Smart Applicationsâ ˘
A˙
I
AND â ˘
AIJMachine Learning for Smart applicationsâ ˘
A˙
I ).
In the initial phase, publications emphasis essentially on
blockchain approaches for smart applications. Then, in the
next phase of the search, we concentrated on an ML-based
solution for blockchain-based smart applications. Based on
the results of these searches, we removed duplicate articles
and obtained our first set of more than 350 publications.
Further, we steered the search procedure with several maga-
zines, journals, and conferences dedicated to the parent field,
ML, smart applications, and Blockchain. Based on that, we
found 130 articles. Then, we studied the different sections
of the articles like abstract, conclusion, and introduction.
Then we categorized these articles as â ˘
AIJrelevantâ ˘
A˙
I or
non-relevantâ ˘
A˙
I to ML for blockchain-based smart applica-
tions. Lastly, we only selected 60 publications to present the
taxonomy, as shown in Figure 3. Here, each layer is color-
coded to tailor each level of the taxonomy. For example, the
root represents the level-0 of the classified taxonomy and
presented in blue color. In a similar way, the four major
dimensions are representing as level-1 in light-green color
and so onwards.
B. TAXONOMY
In this paper, we studied the papers that are predominantly
based on either ML, blockchain, or both. The presented
survey has been divided into four dimensions: Goal Oriented,
Layer Oriented, Countermeasures, and Smart Applications.
Figure 3shows the taxonomy of ML adoption for blockchain-
based smart applications.
IV. GOAL ORIENTED
In this dimension, research work has been included from the
following aspects: (i) preventing, (ii) predicting, (iii) mon-
itoring, (iv) detection, and (v) response of blockchain. For
instance, [29] discussed the detection of intrusion in a collab-
orative manner, [30] predicts bitcoin prices using Bayesian
regression method and [31] monitors the blockchain network
in a distributed manner.
A. DETECTION
This subsection includes the detection of the attacks to
handle data security issues. Meng et al. [12] has proposed
a Collaborative IDS (CIDS) with a data-sharing agreement
on a blockchain-based environment. Data privacy issues of
CIDS can be addressed with the use of ML classifiers on
BT. These classifiers run locally by the data owner, and
the results shared with other users within the network. To
handle the trust computation issue in CIDS, Alexopoulos
et al. [29] proposed an approach to handle the alerts and
data distribution among the participants. Then, it verifies the
data and executes a suitable consensus algorithm to add a
new data block within the blockchain network. These data
alerts could be encrypted with keys distributed to selected
parties. Another approach was to keep them on a separate
blockchain while still being part of the network. Further, an
anomaly detection system (ADS) was implemented based on
the assumption that a similar kind of attack may occur, but on
different nodes in the blockchain network [32]. This system
does not discard information about orphans and forks that
are usually done by other ADS. The attacked nodes shared
this information with other neighborâ ˘
A´
Zs node within the
network. In the experimental analysis, the system success-
fully prevented the same kind of attack (on other nodes) with
negligible overhead.
B. MONITORING
A blockchain monitoring system has been prototyped using
Self Organising Maps (SOMs) [33]. It has modeled the
dataset without external control. Here, large vectors map
to smaller and lower dimension vectors using SOMs. The
system analyzed blockchain data using Kohonen and SOM-
brero libraries (in the R programming environment). The
result shows the effective key attributes monitoring of the
blockchain nodes and finds the emerging patterns. More,
another approach included the distributed pattern recognition
system with the concept of Graph Neurons (GNs) [31] to
monitor the blockchain system. GNs are scalable and could
recognize patterns from similar or incomplete patterns. The
GN communicated with adjacent nodes to detect events
within the network by using input data. Preliminary results
showed object detection by the GN was accurate; still, further
work is needed in the area.
C. PREDICTION
The ML models are mostly used for prediction. A good
prediction model helps to make the right decision making
and analytics. Added to this, an ML model to predict the
bitcoin prices has been proposed by Valenkar et al. [30].
This model uses bayesian regression and random forest with
several features such as block size, total bitcoins, day high,
number of transactions, and trade volume. The trained dataset
was normalized using log, z-score, and box-cox normaliza-
tion techniques. Further, a price prediction study has been
done for several cryptocurrencies such as Ripple, Litecoin,
Dash, Bitcoin, and Ethereum cryptocurrencies [34]. Here,
correlation matrices for feature selections used and reported
the general trends within the network. The proposed model
used multiple regression techniques on bitcoin. The system
showed the 0.9944 accuracies for price prediction.
VOLUME 4, 2016 5
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
Machine Learning Adoption for Blockchain-
based Smart Applications
Smart
Applications
Device
Personalization
Data Trading
Product
Manufacturing
Customer
Services
Counter-
measures
Real-Time
Analytical
Layers
Oriented
User
Process
Endpoint
Application
Layer
Goal Oriented
Prediction
Response
Monitoring
Detection
Prevention UAV
Smart Cities
Energy and
Utilities
FIGURE 3: Proposed Taxonomy for Machine Learning adoption in Blockchain-based Smart Applications
D. RESPONSE
Tsolakis et al. [35] presented an approach for the secure
exchange of energy data between a cluster of end-users and
a Virtual Node. This approach uses a blockchain-based ar-
chitecture to give a demand response solution. This solution
included Fog-Enabled Intelligent Devices (FEID) [36] at the
user side (act as a blockchain node) and managed smart
contracts with the energy producer [36] in cloud computing
platform [37] [38]. Then, a group of energy users combined
into a centralized Virtual Node. All predictions and forecast-
ing about the energy consumption have done on the Virtual
Nodes. The end-user learned itself from each experience that
helps to improve the accuracy of data passed as the next
iteration to the virtual node.
E. PREVENTION
One of the based uses of ML in BT based application is the
prevention of an issue; for instance, employee verification by
a future employer. A mechanism proposed to reduce the time
taken to verify work history details provided by a prospective
employee [39]. The previously worked organization has to
compile details of an employee (date of joining, leaving,
post, etc.), and a public key is generated to encrypt the
data. Subsequently, a new smart contract is created for the
employee. Then, the address of the smart contract is entered
into the organizationâ ˘
A´
Zs database, and later it verified by
the future employer. In a future organization, work history is
fetched from the smart contract of employees and decrypted
using the key. Then, each work history entry compared
with the database of the previous organization. This ensures
consistency and integrity of data as well as the authenticity
of the sender.
A similar idea to tracking studentâ ˘
A´
Zs learning history
has been proposed [40]. The main aim was to maintain the
learning histories of the students and ensure access control,
privacy, and security. To ensure privacy and access con-
trol, three contracts were suggested: Learner â ˘
A¸S Learning
Provider Contract (LLPC), Registrar â ˘
A¸S Learning Provider
Contract (RLPC), and Index Contract (IC). The role of RLPC
is to specify all conditions that describe request access by
the provider within the network. LLPC shows that a learner’s
data exists on the providerâ ˘
A´
Zs database. The IC map the
learners and providers and the corresponding learning his-
tory. Here, specialized Learning Blockchain Application Pro-
gram Interfaces (LB APIs) has been proposed to encourage
the adoption of this technology. Initially, the system is started
with a boot node in the network. A new provider is added
to the network as per the conditions described in RLPC. A
new learner account is created and noted LLPC. In case, the
new provider wants access to learners’ data, it has to request
the learner using the LLPC, and upon granting, data will be
accessible.
6VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
TABLE 2: Comparative Analysis of Goal Oriented Approaches
Authors Year Objective Merits Demerits 1 2 3 4 5
Meng et al. [12] 2018 To present use of blockchain in In-
trusion Detection
Scope of application of blockchain
was discussed
Discusses only data sharing and
trust management issues of col-
labrative intrusion detection
XXXXX
Alexopoulos et al.
[29]
2018 Improvement in trust and account-
ability in Collaborative Intrusion
System using blockchain
Design considerations thoroughly
explained
Use of ML not covered XXXXX
Signorini et al. [32] 2018 to provide a blockchain based
anomaly detection system
extremely less bandwidth overhead
of about 0.248%
Only basic testing carried out con-
sisting of only 3 node
XXXX X
Chawathe et al. [33] 2018 self-organizing maps to monitor
blockchain data in real-time
Running time for the number of
iterations, attributes and instances
shows a linear relationship
Security issues are not highlighted
in storing bitcoin data in a database
XXXXX
Hudaya et al. [31] 2018 distributed pattern recognition
for event monitoring within IoT-
blockchain network
use of graph neuron approach en-
ables distributed pattern recognition
integration of IoT-blockchain not
covered in depth
XXXXX
Saad et al. [34] 2018 predicting bitcoin prices with high
accuracy using ML
user activity dynamics and its ef-
fects discussed extensively
not enough focus on other factors
other than supply demand
XXXXX
Tsolakis et al. [35] 2018 demand response system using
blockchain
security concerns addressed appro-
priately
ML aspects not covered XXXXX
Sarda et al. [39] 2018 To prevent work-history related
frauds using blockchain
Extensive use of blockchain to
share and verify work history
No intersection of blockchain and
ML was found, discussion on secu-
rity issues was not carried out
XXXXX
Liang et al. [41] 2019 Present micro-blockchain based dy-
namic intrusion detection
Scope of application of blockchain
was discussed
Discusses attacks and trust manage-
ment issues only
XXXXX
Sgantzos et al. [42] 2019 Present genetic algorithm imple-
mentation on blockchain
Discussed framework and future
applications
Natural Language interaction with
blockchian based system need to be
explored
XXXXX
Parameters- 1:ML, 2:Blockchain, 3:Detection, 4:Algorithm, 5:Trust Management
Notations- X: Considered, and X: Not Considered.
F. COMPARISON OF EXISTING APPROACHES FOR
GOAL ORIENTED DIMENSION
A detailed comparison of the approaches discussed above
is shown in Table 2. This comparison of works included
several parameters such as trust management, ML, algorithm,
blockchain, detection, merits, and demerits.
V. LAYERS ORIENTED
At this moment, BT is very young, and understanding of
layer division is not enough. Considering the BT as a solitary
layer is similar to enduring everything among the physical
layers and transport layers into a single layer. While separat-
ing the BT into multiple layers, we can understand various
properties of BT which are needed to be implemented. Some
of the properties are: (i) Security: Nodes that do not control
rare resources (commonly computing power) majorly cannot
convince others for a different version of the ledger. (ii) Live-
ness: Here, new blocks can be added to the blockchain with
suitable latency. (iii) Stability: Nodes within the blockchain
network should not amend their belief of the consensus
ledger except rare cases. (iv) Accuracy: Blocks added to the
ledger must signify valid transactions such as they imitate to
a description of how new blocks relate to previous blocks.
This dimension discusses the adoption of ML techniques in
various layers of a blockchain network, such as endpoints,
application layer, process, network (intrusion), and finally,
user level. The proposed categorization into layers recognizes
each property. Security is accomplished at each layer, Live-
ness is realized at the user layer, Stability is achieved at the
network and application layer, and accuracy is achieved at the
endpoints, wherever blocks have significance.
A. APPLICATION LAYER
The application layer helps to accomplish the security, liveli-
ness within the network. An approach to detect malware (in
the form of a portable executable file) has been proposed
using DL methods [43]. The portable executable file was
converted to a grayscale image to feed to the Deep belief
network. In this solution, the receiving node broadcast the
file to other nodes within the network. The file executed
locally in the local detection model of each node. The prob-
ability value of being malware is appended with the block.
A weighted average for the trust of the node is applied to
the appended probabilities. The weighted average calculated
on the condition that the file is malware or not. The model
used Restricted Boltzmann Machines (RBM) techniques with
3000 hidden units. The results showed good accuracy for
malware detection.
B. ENDPOINT
Endpoints included the small computing devices which are
participating in the BT based application network. To handle
the issue of fast computing and reduction in computation
power, emergent technology Edge Computing comes into the
picture. The Edge Computing Service Provider (ECSP) can
meet the requirements, and an approach has been proposed
to maximize the revenue of ECSP [44]. The ECSP allocates
the resource unit to the highest bidder who participated in the
bidding process.
Similarly, Kim et al. [45] proposed an architecture to
offload resources computations in Deep Neural Networks to
edge computing. Within this architecture, the embedded de-
vice (the computation device at the edge server) and the edge
server had to make initial deposits. After that, the embedded
VOLUME 4, 2016 7
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
TABLE 3: Comparative analysis for Layer-Oriented approaches
Author Year Objective 1 2 3 4 5 Pros Cons
Raje et al. [43] 2017 To design a decentralized firewall
using blockchain and deep learning
XXXXXDiscussed Deep Belief Networks
and training procedure in detail and
explored various architectures
Number of devices in experimen-
tal setup was small and working of
blockchain not discussed
Portnoff et al. [46] 2017 To identify human traffickers using
classified ads and bitcoin transac-
tions
XXXXXA detailed discussion on data anal-
ysis
Bitcoin transaction linking was not
perfect, false positives were high,
blockchain aspect not covered
Wasim et al. [47] 2017 Proposed a Law as a service
architecture to monitor Contract
Breaches and issue injunctions
XXXXXResults and system model defined
clearly, proposed unsupervised ML
algorithm
Blockchain not included in the
model description
Luong et al. [44] 2018 Edge computing usage for mining
applications and ML for resource
allocation
XXXXXML algorithms thoroughly dis-
cussed, well-documented results
Focus on optimizing revenue but no
optimal allocation of resources
Kim et al. [45] 2018 Proposed deep neural networks ar-
chitecture for blokchain based edge
computing application
XXXXXExecutes complex programs on
ethereum blockchain using virtual
machines
Experimental results not discussed
in detail, Deep Neural Network ap-
plications not covered
Parameters- 1:Blockchain, 2:Edge Computing, 3:Neural Network, 4:Supervised Learning, 5:Unsupervised Learning
Notations-X: considered, and X: not considered.
device sends the computations to edge servers. Then, the
edge server returns the results, which are verified by other
nodes within the network. The server is rewarded for its
work after validation. The typical Ethereum architecture had
been modified in this implementation by replacing Ethereum
virtual machine by V8 javascript virtual machine. The blocks
generation part had been separated from the virtual machine
to ensure that smart contract execution cannot affect the
rate of generation of blocks. The system performance shows
better results compare to the state-of-art approaches.
C. PROCESS
This subsection comprises ML techniques to the process in-
volves in the blockchain-based applications. An ML classifier
has been used to detect the human traffickers using adult
classified ads [46]. It included ads dataset from Backpage (a
website for online classifieds). It used a supervised learning
model and used logistic regression for classification. The
labeled dataset with identifiers like email and phone number
has been used for testing. This approach takes two ads at a
time as an input to the model. If the ads are from the same
author, the output will be same. The trained model showed an
89.54% true positive rate. Further, this approach used to build
a graph and found links between ads and their related bitcoin
transactions. This approach showed a high false-positive rate.
D. USER
Users are the individual or system that uses the system
functionality seating at one end of the application. Wasim
et. al [47] proposed law as a service ML-based architecture
to monitor contract breaches. Previously, contract breaches
were dealt with or without the use of courts. The pro-
posed system used an unsupervised ML algorithm called
Probability-based Factor Model to issue injunctions. This
model simulated using three service providers Redis [48],
MongoDB [49], and Memcached Servers [50]. Services
monitor the contracts for breaches. The results showed that
services that perform complex operations are more likely to
breach contracts.
E. COMPARISON OF EXISTING APPROACHES FOR
LAYER ORIENTED USE CASES
A detailed comparison of the approaches is discussed in
Table 3. This comparison is made based on several param-
eters such as edge computing, blockchain, neural network,
supervised learning, pros, and cons of the proposed approach.
VI. COUNTERMEASURES
The ML approached in response to a threat or monitoring
a system can be classified as real-time analytics or analysis
based on historical data [33]
A. ANALYTICAL
In todayâ ˘
A´
Zs world, every enterprise generates huge
amounts of data from different sources such as social me-
dia, smartphones, IoT, and other computing devices. These
data are tremendously valuable to organizations. The overall
technique to find a meaningful pattern from these data is
called data analytics. It is a process to convert data from
foresight to insight. It describes what happened in the past,
draw awareness about the present, and make predictions
(with some ML techniques) about the future. ML techniques
can be categories as supervised, unsupervised, and reinforce-
ment learning to analyses the data. The supervised learn-
ing included classification and regression originated on the
idea of example-based learning. Next, unsupervised learning
techniques perform clustering, dimensionality reduction, and
recommendation of a system based on the dataset. This used
to recognize hidden patterns or focus on the well-educated
behavior of the machine. The reinforcement learning ap-
proach helps to reward maximization. In the next decades,
our society will be driven by new technological developments
in ML and BT. Further, a traceability algorithm has been pro-
posed for bitcoin mining by using the ANN approach [51].
It aimed to remove irrelevant data in mining and introduce
traceability to the system. This was helpful in a distributed
architecture and decrease traceability time.
DL is a subset of ML that is originated on definitive
algorithms influenced by the overall structure and work of
the neural network in the human brain. DL rationalizes
tasks to perform speech recognition, image recognition, make
8VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
insightful decisions on natural language processing. DL net-
work consisted of input, output, and hidden multi-layer. It
accepted a new block and a history of the previous block as
an input in BT based applications. It used a state-transition
algorithm on the features such as hash value, nonce, address,
and transaction data. An ML-based classification approach is
used to identify the cyber-crime in Bitcoin [52]. The results
provided a glimpse of the size of cybercrime in the Bitcoin
ecosystem.
B. REAL-TIME
ML enables real-time analytics for all types of data, such as
social â ˘
A¸S accessible, transactional, and operational. It uses
in-memory computing though leaving unremittingly updated
data securely. It improves analytics accuracy and accelerates
the predictive behavior of ML models. It has been noticed
that BT makes real-time cross edge transactions in monetary
and payment frameworks. Several fintech innovators and
banks are currently investigating blockchain due to fast and
real-time settlement of massive amounts independent of geo-
graphic hindrances. Similarly, associations that require real-
time analytics of information on a huge scale can approach
an ML and blockchain-empowered framework to accomplish
the goal. With ML and BT, financial institutions and other
associated organizations across the globe can trace the data
changes to make quick and fast business decisions regard-
less of irregular activities or suspicious transactions. More-
over, the performance of the blockchain-based Software-
Defined Vehicular Network (SDVN) has been improved us-
ing deep Q-learning methodology [53] [54]. This approach
used a permissioned blockchain with Byzantine fault tol-
erance as a consensus protocol. Results showed that this
scheme managed network and computing resources better
and gave the best throughput in the SDVN [55]. Further,
Liu et al. [22] recommended a data collection framework for
Industrial-IoT (IIoT) applications. This framework combines
the use of deep reinforcement learning (DRL) and Ethereum
blockchain. To store and share data, Ethereum nodes were
categories into two categories: mining nodes and nonmining
nodes. The proposed DRL algorithm used three plain compo-
nents, actions (moving path and remoteness of mobile termi-
nals), states (environment description), and rewards (amount
achieved). The proposed algorithm shows the 34.5% increase
in geographical fairness compared to a random solution.
VII. SMART APPLICATIONS
The last dimension covers the adoption of ML techniques
to BT-based smart applications such as data trading, UAV,
product manufacturing, Medical, and Healthcare [56], Smart
Cities, automated Customer Service, and Device Personaliza-
tion as shown in Figure 4. ML and BT are revolutionizing up-
to-date technologies by transforming customer experiences,
behaviors, and business models [57]. Both are making strides
in the major smart application:
FIGURE 4: A list of Smart Applications
A. CUSTOMER SERVICE
With the upsurge of the customer, the Customer service has
to be more efficient and automated to meet rising customer
needs. One of the best ways is to automate the process to
increase a companyâ ˘
A´
Zs capabilities. Wang et al. [58] pro-
posed an AutoML framework for the blockchain-based appli-
cation. It consists six layers: (i) Organization Layer (Includes
entities such as shops, malls, online stores), (ii) Customer
Layer, (iii) Application Layer (applications provide services
to the consumers), (iv) Data conversion layer(Data masking
and unmasking), (v) AutoML layer (Consists various ML
models such as Linear regression, SVM and logistic regres-
sion for predictions), (vi) Blockchain Layer (Secure storage
of data and ensures the safe data exchange) This framework
helps organizations to keep their data safe, automate their
processes and share data with other organizations in a mu-
tually beneficial and safe way.
B. DEVICE PERSONALIZATION
Device Personalization (DP) is a component that uses pre-
dictions across smart devices to improves the quality of
service (QoS) such as actions in the launcher, smart text
selection during writing on text pad. DP Services uses system
permissions to provide smart predictions. In a smart home
environment, and ML model-based single DP framework has
been implemented [59]. Here, smart devices connected to a
smart hub. Every time a user uses a device, a log is generated
with user data, device data, and other parameters for that
device. For example, uses of air conditioner at home. This
log data help to adjust the operation of the device.
C. PRODUCT MANUFACTURING
As a feature of the manufacturing process, organizations
have started trusting blockchain-based procedures to em-
power production, security, transparency, and compliance
checks. Instead of following fixed schedules of machine
VOLUME 4, 2016 9
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
maintenance, ML algorithms are being used to make flexible
plans at specific periods. Product testing and quality control
have also automated increasingly, with versatile algorithms.
It successfully detects faulty and good products, particularly
in profoundly delicate situations. For example, Porsche (a car
manufacturing company) is one of those early adopters of
ML and BT technologies; to improve automobile safety and
increase capabilities. The organization utilizes blockchain
innovation to transfer data more safely and rapidly, offering
its clients peace of mind, regardless of parking, charging, and
third-party access to their car.
D. DATA TRADING
DataTrading is an innovative platform that makes advanced
trading possible for retail traders from all over the world. An
Ethereum based data trading framework has presented that
succeeded in preventing single-point failure and preserving
privacy at the same time [10]. The framework consists of
three entities, data provider, a data consumer, and a market
manager. Once the network is set up, the data consumer and
data provider registered themselves with the market manager.
The data provider needed to deposit an amount with the
manager greater than the amount to be paid by the consumer.
A list indicating topics of data available with the provider is
published. The client referred it and requests some encrypted
data blocks for content validation using a distance metric
learning technique. After successful validation, the client
responded, and the provider sent a signature, and the client
published a smart contract to the network. The provider then
sent another signature, and these two signatures, the client,
decrypted the data. There are different protocols such as setup
protocol, register protocol, payment and acknowledgment
protocol, and query protocol that facilitate the communica-
tion between provider, consumer, and the manager.
E. UNMANNED AERIAL VEHICLE
A UAV or drone is an aircraft that runs without a pilot
(human) aboard. Kuzmin et al. [9] has proposed blockchain-
based UAVNet model includes different devices such as
a network of satellites, cellular base stations, and ground
control stations. Here, BT served multiple purposes, such
as preserving the integrity of the data and for distributed
graphs computation. The communication between satellites
and base stations is prone to electromagnetic jamming, hence
a blockchain-based system enables UAVs to store relevant
coordinateâ ˘
A´
Zs data and operate autonomously within the
jamming zone. This solution used a proof-of-graph consen-
sus algorithm with a simplified memory bounded algorithm
of any existing shortest path to validating a new transaction.
F. SMART CITIES
Smart cities improve the living experiences of individuals.
ML and blockchain emergent technologies play a crucial role
in the innovation of smart cities to provide critical services
and infrastructure components such as healthcare [60], city
administration, smart homes, education, transportation, real
estate, and utilities [61]. (i) Smart Homes: Smart homes
can be monitored, and DP can help to improve the quality
of livelihood. (ii) Smart Parking System: Here, arrival and
departure of vehicles can be tracked for different parking
lots available in a decentralized manner within the smart city
[61]. Consequently, smart parking lots should be designed by
seeing the number of cars in each region. Furthermore, new
parking lots must be recognized spontaneously to benefits in
dealers and vehicle ownerâ ˘
A´
Zs daily life. (iii) Smart Weather
and Water Systems: Here, the system can use some sensors
to generate appropriate data such as temperature, wind speed,
rain, and pressure. The analysis of these data through ML
techniques can contribute to improving the density of smart
cities. (iv) Smart Vehicular Traffic: Vehicular traffic data with
a suitable analysis will benefit the government and citizens to
a great extent [62]. Everyone can decide the arrival time to
a destination by using these data. (v) Surveillance Systems:
Physical security is an utmost important concern for citizens
anywhere they live. To address this issue, smart technology
such as ML and BT can be configured for it. Consequently,
collecting and analyzing data and identifying crimes is one of
the challenging tasks. (vi) Smart Healthcare: It includes the
accessibility of the caregivers and doctors, identification of
nearest medical stores, and clinic. Hence, ML and BT play
a vital role while data is getting generated [63] [64]. (vii)
Smart Governance and Smart Education: Smart governance
can maintain a city smartly. A smart city includes a differ-
ent way of education. It captures the data of the students
and employees in educational and government institutions.
Predictions and analytics required to keep the stuff up to
standards.
G. ENERGY AND UTILITIES
In the Energy industry, BT is assisting to simplify energy
exchanges. For example, IOTA enables smart transformation
across the entire energy industry by implementing BT [65].
It uses the concept of peer-to-peer (P2P) energy production
and consumption. Smart energy microgrids are progressively
developing fame as a technique of making ecological energy
resources. LO3 Energy (a Newyork-based organization) is
also using a blockchain-based development to develop en-
ergy generation, transmission, conservation, and exchanging
within neighborhood networks. The technology uses micro-
grid and smart meters, together with smart contracts, to
manage and track energy transactions. GE Digital, together
with Evolution Energie (a startup), has created an application
that encourages the tracking of renewable energy in SG and
uses blockchain to give energy sources certificate. The idea
is allowing organizations and individuals to trade renewable
energy sources without the involvement of third-party [36].
Globally, most of the industries are dogged by third-party,
who increase the business cost. BT has already interrupted
that model by facilitating the P2P model for customers.
The revolution works even better when combined with the
ML and BT together. We have been witnessed the financial
revolution by using the cryptocurrency world; similarly, we
10 VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
are going to observe more innovation disruptions as a result
of combining ML and BT.
VIII. FUTURE RESEARCH ISSUES AND CHALLENGES
Researchers are looking forward to these technologies across
the globe, but still, various obstacles resist the integration of
BT and ML [66]. Their integration is still in its infancy. Many
open issues and challenges are yet to be addressed. Here,
we discuss the futuristic open issues and challenges of ML
adoption in BT for secure communication, as shown in Figure
5. We highlighted the challenges as Suitability, Infrastructure,
Privacy, security, Memory, Implementation, and Quantum
resilience.
A. SUITABILITY
Blockchain is a viable solution if the source of data cannot be
trusted, and several entities are high in the distributed envi-
ronment. If performance is required, then a simple database
is a better option. Therefore, the architecture of blockchain
must be understood before its use in any application [67].
B. INFRASTRUCTURE
Blockchain specific hardware and network infrastructure im-
prove the performance of many blockchain-based applica-
tions. These could include network administration, mining
hardware, decentralized storage, and communication proto-
cols [30]. However, products tailored to use in blockchain
are still under investigation (involving major tech companies
and financial institutions).
C. PRIVACY
Data generated by devices to be stored on the blockchain is
available to the entire blockchain nodes [33]. This leads to a
potential privacy concern for data that needs to be kept either
private or confidential. Such issues could be resolved by the
use of private blockchains, controlled access, and encryption.
However, the ML models adoption on these limited data
imposed barriers for predictions and analytics.
D. MEMORY
The size of blockchain keeps growing as new blocks are
added to it. Consequently, the entire chain must be stored
by all nodes, which creates a significant memory constraint
on these devices. The performance is also affected by an
increase in chain size. Besides that, the storage of irrelevant
data also wastes computational resources. Blockchains are
immutable, and hence storage management is a major issue
in most implementations.
E. IMPLEMENTATION
A large- scale blockchain network will require an equiva-
lently large number of transactions. Implementation of this
size of blockchain invites potential issues; for instance, high
demand for internet bandwidth, which not easy to reduce,
transactions will be a burden on the network. Hence, the
Research
Challenges
Infrastruct
ure
Memory
Implement
ation
Security
Quantum
resilience
Privacy
Suitability
FIGURE 5: Future Research Issues and Challenges
addition of blocks and transactions needs to be decreased to
meet the inevitable demand.
F. SECURITY
Blockchains are decentralized, and they are prone to security
issues [68]. The most common concern is that due to attacks,
the consensus protocol may be compromised, such that the
mining power of a few farms will control which blocks are
added to the network. This particular concern is present in
public blockchains. Private versions are unaffected by this
attack as they have each node identified, and an appropriate
consensus protocol is in place.
G. QUANTUM RESILIENCE
The hashing algorithms used by blockchains could soon be
broken with quantum computers. A blockchain is predom-
inantly at risk from this because it uses one-way functions
for encryption (only protection in digital signature). This
would cause all the features that make blockchains a viable
storage structure (obsolete). Luckily, quantum computing
offers opportunities to boost the performance and security of
blockchains. Quantum cryptography can reinforce the secu-
rity of the blockchain network as quantum communication is
authenticated inherently (users cannot mimic another user). It
can encrypt entire P2P communications and replace classical
digital signatures in the blockchain network. Research is
underway to design blockchain with quantum computing
[69].
IX. CASE STUDY
In order to demonstrate the proposed architecture for ML
adoption in blockchain-based smart applications, we present
a case study, as shown in Figure 6. In this case study, we study
VOLUME 4, 2016 11
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
Energy Trading Platform
Smart
Grid
Energy Storage
Device
Entities participating in energy trading
Blockchain
Accounts
Industries Buildings
Individual Houses
FIGURE 6: An Energy Trading system
a blockchain-based SG system for energy transactions using
cryptocurrency [70] [71]. This framework uses a blockchain
and DL approach to complete an energy transaction. It is
a reliable P2P energy system. It is based on the Byzantine
fault tolerance algorithm to produce high system throughput.
This approach consists of five phases (i)setup phase, (ii)
agreement phase, (iii) consensus-making phase, (iv) block
creation phase, and (v) a change view phase. Here, blocks
are generated using hash functions and a short signature. It
is an IDS that works on recurrent neural networks to detect
fraudulent transactions and network attacks in blockchain-
based energy applications. The performance of this IDS has
been studied on different energy datasets. It consists of four
entities for communications: (a) Energy Buyer, (b) Energy
Seller, (c) Blockchain, and (d) IDS [41]. Energy buyer trade
energy with the energy seller. An energy buyer shows that
he has sufficient energy money (cryptocurrency) that fulfills
the minimum asset requirement of energy sellers. An energy
buyer will be a person or commercial building or industry.
Energy buyers can be located in a Home Area Network
(HAN), Building Area Network (BAN), or Neighborhood
area Network (NAN). Energy seller entities demonstrate that
it has sufficient energy to sell. Energy sellers can be located
in HAN, BAN, NAN, or energy companies. Energy seller
(entities) produces energy from renewable sources such as
wind energy, solar energy, and biomass. The seller can be
a neighbor, local society with renewable energy resources,
utility provider, or SG. Blockchain entity is a distributed
digital ledger, which is encompassing all energy transactions
in the SG system.
Once an energy seller produces energy from renewable
energy sources and uses that energy for his own use. After
that, if the seller left with energy, then he publishes his per-
unit prices per kilowatt energy on the blockchain network.
In this P2P energy trading system, the needed energy buyer
will look upon the published price and unit of energy. First of
all, the energy buyer checks his account that he has sufficient
cryptocurrency balance for energy trading. Then, if the en-
ergy demand of energy buyers and per-unit price matched his
requirement of energy at that moment, he sends a purchase
request to the energy seller through the blockchain network.
After validation of the buyer on the blockchain network seller
sells the energy to the buyer and receives cryptocurrency in
his account. This P2P energy trading system required further
data analysis to identify the frequent buyer and seller for
the recognition of malicious transactions within the network.
A malicious block or node can impact the P2P system and
compete for energy trading. IDS can help to figure out the
deceitful transactions and network attacks on energy trading
applications [72]. The result shows the good performance of
the system compares to the state-of-art approaches.
12 VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
X. CONCLUSION
The recent advancements in Blockchain and ML have made
them path-breaking technologies. The distributed ledger has
the possibility to work as the backbone of various smart
applications such as smart cities, UAV, SG, data trading. In
this paper, we have presented detailed information on BT
and ML, along with their usages in smart applications and
proposed an ML-BT based architecture. This architecture
can be used to design and deploy an ML-BT based data
analysis system. A discussion and comparison of various
existing surveys are presented. Then, we presented ML-BT
solution taxonomy, focusing on goal oriented, layer oriented,
countermeasures, and smart application dimensions. A com-
parative analysis of available methodologies and approaches
is presented in each dimension. Then, we have listed several
research challenges being faced during ML adoption in BT-
based systems, which require solutions. We also emphasized
a number of research prospects such as infrastructure avail-
ability, quantum resilience, and privacy issues that can serve
as future research directions in this field. Then, we presented
a case study on the energy trading system to verify the
effectiveness of the proposed architecture and concluded the
paper at last.
REFERENCES
[1] S. Kaneriya, J. Vora, S. Tanwar, and S. Tyagi, “Standardising the use of
duplex channels in 5g-wifi networking for ambient assisted living, in
2019 IEEE International Conference on Communications Workshops (ICC
Workshops), pp. 1–6, May 2019.
[2] S. Tyagi, M. S. Obaidat, S. Tanwar, N. Kumar, and M. Lal, “Sensor
cloud based measurement to management system for precise irrigation,”
in GLOBECOM 2017 - 2017 IEEE Global Communications Conference,
pp. 1–6, Dec 2017.
[3] U. Bodkhe, P. Bhattacharya, S. Tanwar, S. Tyagi, N. Kumar, and M. S.
Obaidat, “Blohost: Blockchain enabled smart tourism and hospitality
management,” in 2019 International Conference on Computer, Information
and Telecommunication Systems (CITS), pp. 1–5, Aug 2019.
[4] S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” 2009.
[5] M. Rahouti, K. Xiong, and N. Ghani, “Bitcoin concepts, threats, and
machine-learning security solutions,” IEEE Access, vol. 6, pp. 67189–
67205, 2018.
[6] A. P. Namanya, A. Cullen, I. U. Awan, and J. P. Disso, “The world of
malware: An overview,” in 2018 IEEE 6th International Conference on
Future Internet of Things and Cloud (FiCloud), pp. 420–427, Aug 2018.
[7] S. Krit and E. Haimoud, “Overview of firewalls: Types and policies:
Managing windows embedded firewall programmatically,” in 2017 Inter-
national Conference on Engineering MIS (ICEMIS), pp. 1–7, May 2017.
[8] G. Betarte, E. Gimenez, R. Martinez, and A. Pardo, “Improving web
application firewalls through anomaly detection, in 2018 17th IEEE In-
ternational Conference on Machine Learning and Applications (ICMLA),
pp. 779–784, Dec 2018.
[9] A. Kuzmin and E. Znak, “Blockchain-base structures for a secure and
operate network of semi-autonomous unmanned aerial vehicles,” in 2018
IEEE International Conference on Service Operations and Logistics, and
Informatics (SOLI), pp. 32–37, July 2018.
[10] Y. Zhao, Y. Yu, Y. Li, G. Han, and X. Du, “Machine learning based privacy-
preserving fair data trading in big data market,” Information Sciences,
vol. 478, pp. 449 460, 2019.
[11] C. Pop, T. Cioara, M. Antal, I. Anghel, I. Salomie, and M. Bertoncini,
“Blockchain based decentralized management of demand response pro-
grams in smart energy grids,” Sensors, vol. 18, no. 1, p. 162, 2018.
[12] W. Meng, E. W. Tischhauser, Q. Wang, Y. Wang, and J. Han, “When
intrusion detection meets blockchain technology: A review,” IEEE Access,
vol. 6, pp. 10179–10188, 2018.
[13] M. Conti, E. Sandeep Kumar, C. Lal, and S. Ruj, “A survey on security
and privacy issues of bitcoin, IEEE Communications Surveys Tutorials,
vol. 20, pp. 3416–3452, Fourthquarter 2018.
[14] D. Ucci, L. Aniello, and R. Baldoni, “Survey of machine learning tech-
niques for malware analysis,” Computers & Security, vol. 81, pp. 123
147, 2019.
[15] K. Salah, M. H. U. Rehman, N. Nizamuddin, and A. Al-Fuqaha,
“Blockchain for ai: Review and open research challenges, IEEE Access,
vol. 7, pp. 10127–10149, 2019.
[16] F. Casino, T. K. Dasaklis, and C. Patsakis, “A systematic literature review
of blockchain-based applications: Current status, classification and open
issues,” Telematics and Informatics, vol. 36, pp. 55 81, 2019.
[17] W. Xiong and L. Xiong, “Smart contract based data trading mode using
blockchain and machine learning,” IEEE Access, vol. 7, pp. 102331–
102344, 2019.
[18] T. M. Mitchell, Machine Learning. New York, NY, USA: McGraw-Hill,
Inc., 1 ed., 1997.
[19] P. Louridas and C. Ebert, “Machine learning, IEEE Software, vol. 33,
pp. 110–115, Sep. 2016.
[20] R. Saravanan and P. Sujatha, A state of art techniques on machine learn-
ing algorithms: A perspective of supervised learning approaches in data
classification,” in 2018 Second International Conference on Intelligent
Computing and Control Systems (ICICCS), pp. 945–949, June 2018.
[21] V. Vats, L. Zhang, S. Chatterjee, S. Ahmed, E. Enziama, and K. Tepe, “A
comparative analysis of unsupervised machine techniques for liver disease
prediction,” in 2018 IEEE International Symposium on Signal Processing
and Information Technology (ISSPIT), pp. 486–489, Dec 2018.
[22] C. Liu, X. Xu, and D. Hu, “Multiobjective reinforcement learning: A
comprehensive overview, IEEE Transactions on Systems, Man, and Cy-
bernetics: Systems, vol. 45, pp. 385–398, March 2015.
[23] Z. Zheng, S. Xie, H. Dai, X. Chen, and H. Wang, “An overview of
blockchain technology: Architecture, consensus, and future trends,” in
2017 IEEE International Congress on Big Data (BigData Congress),
pp. 557–564, June 2017.
[24] G. B. Mermer, E. Zeydan, and S. S. Arslan, “An overview of blockchain
technologies: Principles, opportunities and challenges,” in 2018 26th
Signal Processing and Communications Applications Conference (SIU),
pp. 1–4, May 2018.
[25] W. Penard and T. van Werkhoven, “On the secure hash algorithm family,
Cryptography in Context, pp. 1–18, 2008.
[26] M. Vukoli´
c, “The quest for scalable blockchain fabric: Proof-of-work vs.
bft replication,” in Open Problems in Network Security (J. Camenisch
and D. Kesdo˘
gan, eds.), (Cham), pp. 112–125, Springer International
Publishing, 2016.
[27] S. Thompson, P. L. Seijas, and D. Adams, “Scripting smart contracts for
distributed ledger technology. December 2016.
[28] K. Christidis and M. Devetsikiotis, “Blockchains and smart contracts for
the internet of things,” IEEE Access, vol. 4, pp. 2292–2303, 2016.
[29] N. Alexopoulos, E. Vasilomanolakis, N. R. Ivánkó, and M. Mühlhäuser,
“Towards blockchain-based collaborative intrusion detection systems, in
Critical Information Infrastructures Security (G. D’Agostino and A. Scala,
eds.), (Cham), pp. 107–118, Springer International Publishing, 2018.
[30] S. Velankar, S. Valecha, and S. Maji, “Bitcoin price prediction using
machine learning,” in 2018 20th International Conference on Advanced
Communication Technology (ICACT), pp. 144–147, Feb 2018.
[31] A. Hudaya, M. Amin, N. M. Ahmad, and S. Kannan, “Integrating dis-
tributed pattern recognition technique for event monitoring within the iot-
blockchain network,” in 2018 International Conference on Intelligent and
Advanced System (ICIAS), pp. 1–6, Aug 2018.
[32] M. Signorini, M. Pontecorvi, W. Kanoun, and R. D. Pietro, “BAD:
blockchain anomaly detection,” CoRR, vol. abs/1807.03833, 2018.
[33] S. Chawathe, “Monitoring blockchains with self-organizing maps,” in
2018 17th IEEE International Conference On Trust, Security And Privacy
In Computing And Communications/ 12th IEEE International Conference
On Big Data Science And Engineering (TrustCom/BigDataSE), pp. 1870–
1875, Aug 2018.
[34] M. Saad and A. Mohaisen, “Towards characterizing blockchain-based
cryptocurrencies for highly-accurate predictions,” in IEEE INFOCOM
2018 - IEEE Conference on Computer Communications Workshops (IN-
FOCOM WKSHPS), pp. 704–709, April 2018.
[35] A. C. Tsolakis, I. Moschos, K. Votis, D. Ioannidis, T. Dimitrios, P. Pandey,
S. Katsikas, E. Kotsakis, and R. Garcà a-Castro, “A secured and trusted
demand response system based on blockchain technologies,” in 2018
VOLUME 4, 2016 13
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
Innovations in Intelligent Systems and Applications (INISTA), pp. 1–6,
July 2018.
[36] A. Kumari, S. Tanwar, S. Tyagi, N. Kumar, M. S. Obaidat, and J. J.
Rodrigues, “Fog computing for smart grid systems in the 5g environment:
Challenges and solutions,” IEEE Wireless Communications, vol. 26, no. 3,
pp. 47–53, 2019.
[37] J. Vora, P. DevMurari, S. Tanwar, S. Tyagi, N. Kumar, and M. S. Obaidat,
“Blind signatures based secured e-healthcare system,” in 2018 Inter-
national Conference on Computer, Information and Telecommunication
Systems (CITS), pp. 1–5, July 2018.
[38] S. Tanwar, J. Vora, S. Kaneriya, and S. Tyagi, “Fog-based enhanced safety
management system for miners,” in 2017 3rd International Conference on
Advances in Computing,Communication Automation (ICACCA) (Fall),
pp. 1–6, Sep. 2017.
[39] P. Sarda, M. J. M. Chowdhury, A. Colman, M. A. Kabir, and J. Han,
“Blockchain for fraud prevention: A work-history fraud prevention sys-
tem,” in 2018 17th IEEE International Conference On Trust, Secu-
rity And Privacy In Computing And Communications/ 12th IEEE In-
ternational Conference On Big Data Science And Engineering (Trust-
Com/BigDataSE), pp. 1858–1863, Aug 2018.
[40] P. Ocheja, B. Flanagan, and H. Ogata, “Connecting decentralized learning
records: A blockchain based learning analytics platform,” in Proceedings
of the 8th International Conference on Learning Analytics and Knowledge,
LAK ’18, (New York, NY, USA), pp. 265–269, ACM, 2018.
[41] H. Liang, J. Wu, S. Mumtaz, J. Li, X. Lin, and M. Wen, “Mbid: Micro-
blockchain-based geographical dynamic intrusion detection for v2x,”
IEEE Communications Magazine, vol. 57, no. 10, pp. 77–83, 2019.
[42] K. Sgantzos and I. Grigg, “Artificial intelligence implementations on the
blockchain. use cases and future applications,” Future Internet, vol. 11,
no. 8, p. 170, 2019.
[43] S. Raje, S. Vaderia, N. Wilson, and R. Panigrahi, “Decentralised firewall
for malware detection,” in 2017 International Conference on Advances in
Computing, Communication and Control (ICAC3), pp. 1–5, Dec 2017.
[44] N. C. Luong, Z. Xiong, P. Wang, and D. Niyato, “Optimal auction for
edge computing resource management in mobile blockchain networks:
A deep learning approach,” in 2018 IEEE International Conference on
Communications (ICC), pp. 1–6, May 2018.
[45] J.-Y. Kim and S.-M. Moon, “Blockchain-based edge computing for deep
neural network applications,” in Proceedings of the Workshop on INTel-
ligent Embedded Systems Architectures and Applications, INTESA ’18,
(New York, NY, USA), pp. 53–55, ACM, 2018.
[46] R. S. Portnoff, D. Y. Huang, P. Doerfler, S. Afroz, and D. McCoy, “Back-
page and bitcoin: Uncovering human traffickers, in Proceedings of the
23rd ACM SIGKDD International Conference on Knowledge Discovery
and Data Mining, KDD ’17, (New York, NY, USA), pp. 1595–1604, ACM,
2017.
[47] M. U. Wasim, A. A. Z. A. Ibrahim, P. Bouvry, and T. Limba, “Law as
a service (laas): Enabling legal protection over a blockchain network, in
2017 14th International Conference on Smart Cities: Improving Quality of
Life Using ICT IoT (HONET-ICT), pp. 110–114, Oct 2017.
[48] “redis.” https://redis.io/. Online; Accessed: 4-December-2019.
[49] “mongodb.” https://www.mongodb.com/. Online; Accessed: 4-November-
2019.
[50] “Memcached.” https://memcached.org/. Online; Accessed: 11-November-
2019.
[51] R.-Y. Chen, A traceability chain algorithm for artificial neural networks
using ˘
A¸Ss fuzzy cognitive maps in blockchain, Future Generation
Computer Systems, vol. 80, pp. 198 210, 2018.
[52] H. Sun Yin and R. Vatrapu, “A first estimation of the proportion of
cybercriminal entities in the bitcoin ecosystem using supervised machine
learning,” in 2017 IEEE International Conference on Big Data (Big Data),
pp. 3690–3699, Dec 2017.
[53] C. Qiu, F. R. Yu, F. Xu, H. Yao, and C. Zhao, “Blockchain-based
distributed software-defined vehicular networks via deep q-learning, in
Proceedings of the 8th ACM Symposium on Design and Analysis of In-
telligent Vehicular Networks and Applications, DIVANet’18, (New York,
NY, USA), pp. 8–14, ACM, 2018.
[54] J. Bhatia, R. Dave, H. Bhayani, S. Tanwar, and A. Nayyar, “Sdn-based
real-time urban traffic analysis in vanet environment, Computer Commu-
nications, vol. 149, pp. 162–175, 10 2019.
[55] J. Bhatia, Y. Modi, S. Tanwar, and M. Bhavsar, “Software defined vehicu-
lar networks: A comprehensive review, International Journal of Commu-
nication Systems, vol. 32, no. 12, p. e4005, 2019.
[56] J. Vora, A. Nayyar, S. Tanwar, S. Tyagi, N. Kumar, M. S. Obaidat,
and J. J. P. C. Rodrigues, “Bheem: A blockchain-based framework for
securing electronic health records,” in 2018 IEEE Globecom Workshops
(GC Wkshps), pp. 1–6, Dec 2018.
[57] S. K. Singh, M. M. Salim, M. Cho, J. Cha, Y. Pan, and J. H. Park, “Smart
contract-based pool hopping attack prevention for blockchain networks,
Symmetry, vol. 11, no. 7, p. 941, 2019.
[58] W. M. Wang, H. Guo, Z. Li, Y. Shen, and A. V. Barenji, “Towards open
and automated customer service: A blockchain-based automl framework,
in Proceedings of the 2Nd International Conference on Computer Science
and Application Engineering, CSAE ’18, (New York, NY, USA), pp. 27:1–
27:6, ACM, 2018.
[59] K. Singla, J. Bose, and S. Katariya, “Machine learning for secure device
personalization using blockchain,” in 2018 International Conference on
Advances in Computing, Communications and Informatics (ICACCI),
pp. 67–73, Sep. 2018.
[60] R. Gupta, S. Tanwar, S. Tyagi, N. Kumar, M. S. Obaidat, and B. Sadoun,
“Habits: Blockchain-based telesurgery framework for healthcare 4.0, in
2019 International Conference on Computer, Information and Telecom-
munication Systems (CITS), pp. 1–5, Aug 2019.
[61] A. Kumari, S. Tanwar, S. Tyagi, N. Kumar, M. Maasberg, and K.-K. R.
Choo, “Multimedia big data computing and internet of things applications:
A taxonomy and process model,” Journal of Network and Computer
Applications, vol. 124, pp. 169–195, 2018.
[62] A. Mewada, S. Tanwar, and Z. Narmawala, “Comparison and evaluation
of real time reservation technologies in the intelligent public transport
system,” in 2018 Fifth International Conference on Parallel, Distributed
and Grid Computing (PDGC), pp. 800–805, Dec 2018.
[63] A. Kumari, S. Tanwar, S. Tyagi, and N. Kumar, “Fog computing for
healthcare 4.0 environment: Opportunities and challenges, Computers &
Electrical Engineering, vol. 72, pp. 1–13, 2018.
[64] S. Tanwar, K. Parekh, and R. Evans, “Blockchain-based electronic health-
care record system for healthcare 4.0 applications,” Journal of Information
Security and Applications, vol. 50, p. 102407, 2020.
[65] “Iota.” https://www.iota.org/. Online; Accessed: 4-November-2019.
[66] S. K. Singh, S. Rathore, and J. H. Park, “Blockiotintelligence: A
blockchain-enabled intelligent iot architecture with artificial intelligence,”
Future Generation Computer Systems, 2019.
[67] K. Salah, M. H. U. Rehman, N. Nizamuddin, and A. Al-Fuqaha,
“Blockchain for ai: review and open research challenges, IEEE Access,
vol. 7, pp. 10127–10149, 2019.
[68] A. Kumari, S. Tanwar, S. Tyagi, and N. Kumar, “Verificationand validation
techniques for streaming big data analytics in internet of things environ-
ment,” IET Networks, vol. 8, no. 2, pp. 92–100, 2018.
[69] A. K. Fedorov, E. O. Kiktenko, and A. I. Lvovsky, “Quantum computers
put blockchain security at risk,” 2018.
[70] M. A. Ferrag and L. Maglaras, “DeepCoin: A Novel Deep Learning and
Blockchain-Based Energy Exchange Framework for Smart Grids, IEEE
Transactions on Engineering Management, vol. PP, pp. 1–13, 2019.
[71] N. Wang, X. Zhou, X. Lu, Z. Guan, L. Wu, X. Du, and M. Guizani, “When
energy trading meets blockchain in electrical power system: The state of
the art,” Applied Sciences, vol. 9, no. 8, p. 1561, 2019.
[72] N. Kabra, P. Bhattacharya, S. Tanwar, and S. Tyagi, “Mudrachain:
Blockchain-based framework for automated cheque clearance in financial
institutions,” Future Generation Computer Systems, vol. 102, pp. 574
587, 2020.
14 VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2961372, IEEE Access
Tanwar et al.: Machine Learning adoption in Blockchain-based Smart Applications: The challenges, and a way forward
SUDEEP TANWAR is an Associate Professor
in the Computer Science and Engineering De-
partment at Institute of Technology, Nirma Uni-
versity, Ahmedabad, Gujarat, India. He is visit-
ing Professor in Jan Wyzykowski University in
Polkowice, Poland and University of Pitesti in
Pitesti, Romania. He received B.Tech in 2002 from
Kurukshetra University, India, M.Tech (Honor’s)
in 2009 from Guru Gobind Singh Indraprastha
University, Delhi, India and Ph.D. in 2016 with
specialization in Wireless Sensor Network. He has authored or coauthored
more than 100 technical research papers published in leading journals
and conferences from the IEEE, Elsevier, Springer, Wiley, etc. Some of
his research findings are published in top cited journals such as IEEE
Transactions on TVT, IEEE Transactions on Industrial Informatics, Applied
Soft Computing, Journal of Network and Computer Application, Pervasive
and Mobile Computing, International Journal of Communication System,
Telecommunication System, Computer and Electrical Engineering and IEEE
Systems Journal. He has also published three edited/authored books with
International/National Publishers. He has guided many students leading to
M.E./M.Tech and guiding students leading to Ph.D. He is Associate Editor
of IJCS, Wiley and Security and Privacy Journal, Wiley. His current interest
includes Wireless Sensor Networks, Fog Computing, Smart Grid, IoT, and
Blockchain Technology. He was invited as Guest Editors/Editorial Board
Members of many International Journals, invited for keynote Speaker in
many International Conferences held in Asia and invited as Program Chair,
Publications Chair, Publicity Chair, and Session Chair in many International
Conferences held in North America, Europe, Asia and Africa. He has been
awarded best research paper awards from IEEE GLOBECOM 2018, IEEE
ICC 2019, and Springer ICRIC-2019.
QASIM BHATIA is a graduate student of Nirma
University, Ahmedabad, India. His research inter-
est includes machine learning, network security,
fog computing, and cloud computing.
PRUTHVI PATEL is a graduate student of Nirma
University, Ahmedabad, India. His research inter-
est includes big data analytics, fog computing, and
cloud computing.
APARNA KUMARI pursuing Ph.D. from Depart-
ment of Computer Science and Engineering, In-
stitute of Technology, Nirma University, Ahmed-
abad, Gujarat, India. She received M.Tech from
Jawaharlal Nehru University, Delhi, India in 2012.
Her research interest includes big data analytics,
smart grid, blockchain technology and cloud com-
puting.
PRADEEP KUMAR SINGH is currently work-
ing as Assistant Professor (Senior Grade) in De-
partment of CSE at Jaypee University of Informa-
tion Technology (JUIT), Waknaghat, H.P. He has
completed his Ph.D. in Computer Science & En-
gineering from Gautam Buddha University (State
Government University), Greater Noida, UP, In-
dia. He received his M.Tech. (CSE) with Distinc-
tion from GGSIPU , New Delhi, India. Dr. Singh
is having life membership of Computer Society of
India (CSI), Life Member of IEI and promoted to Senior Member Grade
from CSI and ACM. He is Associate Editor of International Journal of
Information Security and Cybercrime (IJISC) a scientific peer reviewed
journal from Romania. He has published nearly 85 research papers in various
International Journals and Conferences of repute. He has received three
sponsored research projects grant from Govt. of India and Govt. of HP worth
Rs 25 Lakhs. He has edited total 8 books from Springer and Elsevier and
also edited several special issues for SCI and SCIE Journals from Elsevier
and IGI Global. He has Google scholar citations 401, H-index 12 and i-10
Index 15 in his account.
WEI-CHIANG HONG (M’04-SM’10) is a Pro-
fessor in the Department of Information Manage-
ment, Oriental Institute of Technology, Taiwan.
His research interests mainly include computa-
tional intelligence (neural networks and evolution-
ary computation), and application of forecasting
technology (ARIMA, support vector regression,
and chaos theory) and machine learning algo-
rithms. Professor Hong serves as the program
committee of various international conferences
including premium ones such as IEEE CEC, IEEE CIS, IEEE ICNSC,
IEEE SMC, IEEE CASE, and IEEE SMCia, etc. In May 2012, his paper
had been evaluated as Top Cited Article 2007-2011 by Elsevier Publisher
(Netherlands). In Sep. 2012, once again, his paper had been indexed in
ISI Essential Science Indicator database as Highly Cited Papers, in the
meanwhile, he also had been awarded as the Model Teacher Award by
Taiwan Private Education Association. Professor Hong is a senior member of
IIE and IEEE. He is indexed in the list of Who’s Who in the World (25th-30th
Editions), Who’s Who in Asia (2nd Edition), and Who’s Who in Science and
Engineering (10th and 11th Editions). Professor Hong is currently appointed
as the Editor-in-Chief of the International Journal of Applied Evolutionary
Computation, in addition, he serves as a guest editor of the Energies, and
is appointed as an Associate Editor of the Neurocomputing, Forecasting,
and International Journal of System Dynamics Applications. He has Google
scholar citations 5424, H-index 38 and i-10 Index 64 in his account..
VOLUME 4, 2016 15
... The convergence of blockchain technology and machine learning (ML) represents a significant milestone in the evolution of smart applications, offering a new paradigm for addressing some of the most pressing challenges in data security, privacy, and system efficiency [1]. As both blockchain and ML technologies have matured independently, their integration holds the potential to create a more robust and intelligent infrastructure for various applications, ranging from finance and healthcare to IoT and supply chain management [2]. ...
... As ML models become more complex and dataintensive, ensuring that blockchain systems can scale to accommodate these demands is crucial [11]. Moreover, the interoperability between different blockchain networks and ML frameworks remains an area of ongoing research, as seamless integration is necessary to unlock the full potential of these combined technologies [2]. This paper aims to provide a comprehensive exploration of the adoption of machine learning in blockchain-based smart applications, with a particular focus on the perspectives of security and efficiency [12]. ...
... In the realm of IoT, blockchain and ML offer solutions for managing the vast amounts of data generated by connected devices. For example, in a smart grid, ML algorithms can optimize energy usage based on real-time data, while blockchain ensures the security and transparency of energy transactions [2]. Additionally, blockchain-enhanced ML can improve supply chain management by providing real-time tracking of goods and ensuring the authenticity of transactions [14]. ...
Article
This study investigates the integration of machine learning (ML) into blockchain-based smart applications, aiming to enhance security, efficiency, and scalability. The research contributes a novel framework that combines blockchain's decentralized ledger with privacy-preserving ML techniques, addressing key challenges in data integrity and computational efficiency. The primary objective is to evaluate the performance of this integration in a simulated smart grid environment, focusing on security, processing time, energy consumption, and scalability. Our findings reveal that the integrated system significantly improves security, achieving a 98% success rate in mitigating data breaches and reducing the impact of adversarial attacks by 90%. Computational efficiency is also enhanced, with the optimized blockchain-ML configuration reducing processing time by 33% and energy consumption by 20% compared to standard blockchain setups. However, scalability remains a challenge; the system demonstrates effective scalability up to 100 nodes, beyond which transaction processing time increases by 50%, indicating the need for further optimization. The results suggest that while the integration of ML and blockchain offers substantial improvements in security and efficiency, addressing scalability and environmental impact are critical for broader application. The novelty of this research lies in its dual focus on enhancing both security and efficiency within blockchain-ML systems, providing a foundation for future advancements in decentralized intelligent applications across industries. This work contributes to the field by offering empirical data that supports the viability of blockchain-ML integration and by highlighting the areas where further research is needed to realize its full potential.
... Machine learning (ML) can optimize scholarship fund allocation by leveraging its advanced learning capabilities to identify deserving students. Combining blockchain with ML has significant potential to create a secure, decentralized, and efficient administrative system [8], [9]. This study investigates the possibility of integrating blockchain and ML within our innovative platform to predict scholarship qualification effectively. ...
Article
Full-text available
Traditional scholarship management systems are often marred by inefficiencies, a lack of transparency, and significant concerns regarding data privacy. These challenges hinder the equitable distribution of educational funds and obscure the scholarship allocation process. This study aims to address these issues by proposing a blockchain-based scholarship management platform. However, scalability issues are introduced along with blockchain-based solutions, for which zk-rollup offers a promising solution. The integration of Self-Sovereign Identity Zero-Knowledge Proof assures secure and private submission of scholarship applications, safeguarding student data while maintaining transparency in the verification process. Additionally, a machine learning model is employed to predict scholarship qualification, functioning as a recommendation system that identifies and prioritizes deserving students. This data-driven approach proactively eliminates barriers that potentially impede deserving students from accessing financial aid, such as administrative oversights or a lack of self-assurance in their qualifications. Our experimental findings confirm the effectiveness of zk-rollups in enhancing transaction efficiency, demonstrating a reduction in transaction confirmation time by approximately 63.6% and a decrease in transaction costs by nearly 90%. Besides, the machine learning model achieved a good performance rating, achieving a balanced accuracy of 86.75% and a mean average precision of 91.68% on a realistically imbalanced test set, reflecting real-world conditions.
... Machine learning enabled block chain technologies have revolutionized the way vehicles are maintained and serviced. These technologies combine the power of machine learning algorithms with the security and transparency of block chain to create efficient and reliable maintenance strategies for vehicles [42]. By analyzing large amounts of data, machine learning algorithms can identify patterns and trends that can be used to predict maintenance needs and optimize vehicle performance. ...
Article
Full-text available
The traditional approach to vehicle maintenance in the automotive industry is often reactive, leading to increased downtime, higher costs, and decreased productivity. There is a need for a more proactive and data-driven approach to maintenance that can help identify potential issues before they escalate. Machine learning offers the potential to analyze vast amounts of data and predict maintenance needs accurately, leading to more efficient operations. To investigate the adoption of machine learning in streamlining vehicle maintenance strategies, a comprehensive literature review was conducted to understand the current state of the automotive industry and the potential benefits of machine learning in maintenance operations. Case studies and examples of companies that have successfully implemented machine learning in their maintenance strategies were also analyzed to identify best practices. The study revealed that machine learning can help automotive companies optimize their maintenance schedules, prioritize critical maintenance tasks, and improve the overall reliability of their vehicles. Consequently, enabling these companies stay competitive in a rapidly changing market by supporting them to quickly adapt to new technologies and customer demands. This proactive approach to maintenance is observed as a viable tool that can prevent costly breakdowns and reduce downtime, ultimately leading to increased productivity and profitability. However, from the findings of this study, adoption of machine learning in vehicle maintenance strategies is still in its early stages within the automotive industry. While some companies have commence implementation, many are still hesitant to fully embrace this technology. Barriers to adoption include concerns about data security, lack of expertise in machine learning, and resistance to change within organizations. With the conventional trends in vehicle maintenance strategies, it is essential for automotive companies to stay ahead of the curve and leverage this technology to drive innovation and success in their operations.
... This review paper reveals a significant uptick in research activity focused on the integration of blockchain technology within IoT-based healthcare systems for security and privacy purposes, with a notable acceleration observed since 2019. Tanwar et al. [54] divided the application of blockchain technology, from a security perspective, into five main categories in their review study (detection, monitoring, prediction, response, and prevention). In a survey, Panarello et al. [55] divided studies that focused on blockchain integration with IoT into five categories according to the security concept they addressed. ...
Article
Full-text available
The Internet of Medical Things (IoMT) is a rapidly expanding network comprising medical devices, sensors, and software that collect and exchange patient health data. Today, the IoMT has the potential to revolutionize healthcare by offering more personalized care to patients and improving the efficiency of healthcare delivery. However, the IoMT also introduces significant privacy concerns, particularly regarding data privacy. IoMT devices often collect and store large amounts of data about patients’ health. These data could be used to track patients’ movements, monitor their health habits, and even predict their future health risks. This extensive data collection and surveillance could be a major invasion of patient privacy. Thus, privacy-preserving research in an IoMT context is an important area of research that aims to mitigate these privacy issues. This review paper comprehensively applies the PRISMA methodology to analyze, review, classify, and compare current approaches of preserving patient data privacy within IoMT blockchain-based healthcare environments.
Article
Securing sensitive data in cloud storage systems is crucial with the increasing sophistication of cyber-attacks in today’s digital ecosystem. The rising need to protect cloud storage systems from illegal access, data breaches, and new cyber threats makes the research necessary. Because of the difficulty in keeping up with the ever-changing nature of security threats and user behaviour, innovative, flexible, and dynamic ways to access management have had to be developed. The ever-changing nature of modern threats and the requirement for constant monitoring and instantaneous responses provide difficulties for cloud storage security. The present research presents Deep Reinforcement Learning-Enhanced Adaptive Access Control (DRL-AAC), which makes access control decisions that are dynamic and aware of their context using modern reinforcement learning algorithms. It has been used to design and implement a Deep Reinforcement Learning (DRL-AAC) enhanced adaptive access control system for dynamically managing cloud storage. The system should be able to make access determinations in real-time, adapting to changing security threats and user habits. DRL-AAC uses user profiling to analyze and detect abnormalities and unauthorized access based on user behaviour patterns. DRL-AAC has wide-ranging potential uses, from consumer to enterprise to government cloud storage. Because of its intelligence and flexibility, it may be used to protect data in various settings. This research provides a comprehensive simulation analysis to prove the efficiency of DRL-AAC. The experimental findings demonstrate that the suggested DRL-AAC model increases the computational intensity analysis ratio of 96.2%, level of data privacy of 97.8%, adaptability ratio of 95.9%, resource allocation efficiency of 97.5% efficiency, scalability ratio of 98.8% compared to other existing models. The purpose of the research is to demonstrate the efficacy of DRL-AAC in doing its intended task of detecting and neutralizing potential security risks with as few false positives and negatives as possible. The paper aspires to verify DRL-AAC as a ground-breaking approach to bolstering cloud storage security. It can potentially improve cloud storage security by responding in real-time to new threats and user habits, protecting sensitive information on the cloud. The proposed work is evaluated by investigating its computational intensity, data privacy, adaptability, resource efficiency, and scalability.
Chapter
Over the last few years, there has been a significant increase in the amount of research conducted in the area of blockchain technology, which is a relatively new technology that has captivated the interest of the general public. At the same time, another domain had a significant data growing volume helped the widespread use of machine learning. One important element to create an efficient machine learning model is the collected ground truth data. This data can be misinterpreted as the golden truth based on the collection approach. Therefore, we are proposing that the blockchain can significantly facilitate gathering the ground truth for machine learning. The main goal of this paper is to identify several important aspects of integrating blockchain and crowdsourcing as an approach that can be utilized to collect ground truth data. Based on the literature research, we identified 11 blockchain. Following, we approached experts to collect online questioner answers, the questioner aimed to determine the blockchain characteristics with an impact toward the crowdsourcing and ground truth.
Article
Blockchain provides several advantages, including decentralization, data integrity, traceability, and immutability. However, despite its advantages, blockchain suffers from significant limitations, including scalability, resource greediness, governance complexity, and some security related issues. These limitations prevent its adoption in mainstream applications. Artificial Intelligence (AI) can help addressing some of these limitations. This survey provides a detailed overview of the different blockchain AI-based optimization and improvement approaches, tools and methodologies proposed to meet the needs of existing systems and applications with their benefits and drawbacks. Afterwards, the focus is on suggesting AI-based directions where to address some of the fundamental limitations of blockchain.
Article
Full-text available
Smart housing, therefore, theoretically becomes very vital in this context of a smart city for sustainable urban planning and development. Machine learning technologies can be considered quite fundamental in enhancing efficiency, sustainability, and livability through incorporating into smart housing. However, rapid urbanization, population growth, traffic congestion, and energy management are huge problems. The main objective of this research work is to identify the feasibility of ML application in smart housing for resource management optimization, environmental sustainability, and public safety. It conducts an analysis on key factors like energy consumption, waste management, and public safety measures by applying machine learning’s efficient algorithms on the comprehensive dataset. There is a 20% decrease in total energy consumption, 15% increase in renewable source energy consumption, and a 25% efficiency improvement in waste management. In addition, public safety response times decreased by 30%. Also, ML models gave out very accurate predictions for power use, traffic patterns, and air quality that turned out with an average accuracy of 92%, thus saving 10% carbon emissions. The study clearly showed that ML will play a very key role in housing planning and interior design. The results bring out the importance of ML in tackling challenging urban issues and promoting better sustainable urban planning practices.
Article
Full-text available
Anomaly detection tools play a role of paramount importance in protecting networks and systems from unforeseen attacks, usually by automatically recognizing and filtering out anomalous activities. Over the years, different approaches have been designed, all focused on lowering the false positive rate. However, no proposal has addressed attacks specifically targeting blockchain-based systems. In this paper, we present BAD: Blockchain Anomaly Detection. This is the first solution, to the best of our knowledge, that is tailored to detect anomalies in blockchain-based systems. BAD is a complete framework, relying on several components leveraging, at its core, blockchain meta-data in order to collect potentially malicious activities. BAD enjoys some unique features: (i) it is distributed (thus avoiding any central point of failure); (ii) it is tamper-proof (making it impossible for a malicious software to remove or to alter its own traces); (iii) it is trusted (any behavioral data is collected and verified by the majority of the network); and, (iv) it is private (avoiding any third party to collect/analyze/store sensitive information). Our proposal is described in detail and validated via both experimental results and analysis, that highlight the quality and viability of our Blockchain Anomaly Detection solution.
Article
Full-text available
An exemplary paradigm of how an AI can be a disruptive technological paragon via the utilization of blockchain comes straight from the world of deep learning. Data scientists have long struggled to maintain the quality of a dataset for machine learning by an AI entity. Datasets can be very expensive to purchase, as, depending on both the proper selection of the elements and the homogeneity of the data contained within, constructing and maintaining the integrity of a dataset is difficult. Blockchain as a highly secure storage medium presents a technological quantum leap in maintaining data integrity. Furthermore, blockchain’s immutability constructs a fruitful environment for creating high quality, permanent and growing datasets for deep learning. The combination of AI and blockchain could impact fields like Internet of things (IoT), identity, financial markets, civil governance, smart cities, small communities, supply chains, personalized medicine and other fields, and thereby deliver benefits to many people.
Article
Full-text available
Pool hopping attack is the result of miners leaving the pool when it offers fewer financial rewards and joining back when the rewards of mining yield higher rewards in blockchain networks. This act of leaving and rejoining the pool only during the good times results in the miner receiving more rewards than the computational power they contribute. Miners exiting the pool deprive it of its collective hash power, which leaves the pool unable to mine the block successfully. This results in its competitors mining the block before they can finish mining. Existing research shows pool hopping resistant measures and detection strategies; however, they do not offer any robust preventive solution to discourage miners from leaving the mining pool. To prevent pool hopping attacks, a smart contract-based pool hopping attack prevention model is proposed. The main objective of our research is maintaining the symmetrical relationship between the miners by requiring them all to continually contribute their computational power to successfully mine a block. We implement a ledger containing records of all miners, in the form of a miner certificate, which tracks the history of the miner's earlier behavior. The certificate enables a pool manager to better initiate terms of the smart contract, which safeguards the interests of existing mining pool members. The model prevents frequent mine hoppers from pool hopping as they submit coins in the form of an escrow and risk losing them if they abandon the pool before completing mining of the block. The key critical factors that every pool hopping attack prevention solution must address and a study of comparative analysis with existing solutions are presented in the paper.
Article
Full-text available
There are two traditional data trading modes, the hosting mode and the aggregation mode, which depend on the trusted third parties to a large extent. The hosting mode is that the data is completely hosted in the data trading center, so the data trading center retains the data. On the surface, the aggregation mode is that the data trading center is not to retain the data of trading, but actually it has the ability to retain the data. There is a fundamental difference between the ability to retain the data and the inability to retain the data. These two trading modes cause the data owners to be afraid to share the data trading. In this paper, we propose a solution of the data trading mode based on smart contract using blockchain and machine learning. Our solution takes advantage of the immutability, tamper-proof and traceability of blockchain, the programmability of smart contract, and the verification of data availability by the similarity learning to propose a challenge response mechanism between the data purchaser and the data owner, an off-chain download mechanism between the data purchaser and the data storage service provider, and an arbitration mechanism for the controversy resolution of the data trading. The challenge response mechanism is used to authenticate and authorize the data owner, the off-chain download mechanism is used to authenticate and authorize the data purchaser to download the purchased data, and the similarity learning is used to deal with the controversy over the data availability in the data trading. The design and implementation of data trading smart contract successfully achieved the goal of removing the trusted third party in the data trading, thus the problem that the data trading center has the ability to retain the data in the process of the data trading is solved, as well as the automatic payment by using the Ethereum encrypted currency among the trading participants is realized. This paper presents the whole process of smart contract from the design, implementation to the test completion, and provides the security analysis and performance evaluation. The full code of smart contract and the ABI interface have been uploaded to the GitHub for the public release.
Article
Modern healthcare systems are characterized as being highly complex and costly. However, this can be reduced through improved health record management, utilization of insurance agencies, and blockchain technology. Blockchain was first introduced to provide distributed records of money-related exchanges that were not dependent on centralized authorities or financial institutions. Breakthroughs in blockchain technology have led to improved transactions involving medical records, insurance billing, and smart con- tracts, enabling permanent access to and security of data, as well as providing a distributed database of transactions. One significant advantage of using blockchain technology in the healthcare industry is that it can reform the interoperability of healthcare databases, providing increased access to patient medical records, device tracking, prescription databases, and hospital assets, including the complete life cycle of a device within the blockchain infrastructure. Access to patients’ medical histories is essential to cor- rectly prescribe medication, with blockchain being able to dramatically enhance the healthcare services framework. In this paper, several solutions for improving current limitations in healthcare systems using blockchain technology are explored, including frameworks and tools to measure the performance of such systems, e.g., Hyperledger Fabric, Composer, Docker Container, Hyperledger Caliper, and the Wireshark capture engine. Further, this paper proposes an Access Control Policy Algorithm for improving data ac- cessibility between healthcare providers, assisting in the simulation of environments to implement the Hyperledger-based eletronic healthcare record (EHR) sharing system that uses the concept of a chain- code. Performance metrics in blockchain networks, such as latency, throughput, Round Trip Time (RTT). have also been optimized for achieving enhanced results. Compared to traditional EHR systems, which use client-server architecture, the proposed system uses blockchain for improving efficiency and security.
Article
Vehicle-to-everything (V2X) aims to make transportation system more intelligent through linking everything with the moving vehicles, but it brings geographical dynamic intrusions. However, existing intrusion detection systems (IDSs) of vehicles just deploy the preset statics strategies. As a novel security technology, blockchain can realize decentralized tamper- resistance. However, it has not been used for IDSs because of its rigid structure. In this article, we propose Micro-Blockchain based geographical dynamic Intrusion Detection, MBID, for V2X. A novel nested micro-blockchain structure is proposed, where each micro-blockchain deployed in a small region can construct local intrusion detection strategies for vehicles with tamper- resistance. When a vehicle moves to another region, spatial-temporal dynamic IDSs strategies are constructed through the proposed repeatedly nested scheme for micro-blockchains. Moreover, the control plane is proposed to dynamically configure IDSs strategies into the micro-blockchain. Simulation results show the accuracy of MBID.
Article
Accurate and real-time traffic flow prediction plays a central role for efficient traffic management. Software Defined Networking (SDN) is one of the key concerns in networking that has caught much attention in recent years. Extensive advancement in software-based configurable equipment has made ready for a new networking paradigm called software-defined vehicular networks (SDVNs). In this paper, we propose a data-driven approach for implementing an artificially intelligent model for vehicular traffic behavior prediction. We combine the flexibility, scalability, and adaptability leveraged by the SDVN architecture along with the machine learning algorithms to model the traffic flow efficiently. First, we introduce an ingenious approach to find congestion sensitive spots in the VANET by means of clustering algorithm and then predicting the future traffic densities for each spot by recurrent neural networks (RNNs). Neural networks (NNs) have been extensively used to model short-term traffic prediction in the past years. We construct a long short-term memory neural network (LSTM-NN) architecture which overcomes the issue of back-propagated error decay through memory blocks for spatiotemporal traffic prediction with high temporal dependency. Due to such learning ability, LSTM can capture the stochastic characteristics and non-linear nature of the traffic flow. The performance evaluation of the proposed approach shows that it has the potential to predict real-time traffic trends accurately with Mean Squared Error (MSE) of 3.0e − 3 in the scaled metrics, which is equivalent to 97% accuracy in traffic density prediction.
Article
In the recent year, Internet of Things (IoT) is industrializing in several real-world applications such as smart transportation, smart city to make human life reliable. With the increasing industrialization inIoT, an excessive amount of sensing data is producing from various sensors devices in the Industrial IoT. To analyzes of big data, Artificial Intelligence (AI) plays a significant role as a strong analytic tool and delivers a scalable and accurate analysis of data in real-time. However, the design and development of a useful big data analysis tool using AI have some challenges, such as centralized architecture, security, and privacy, resource constraints, lack of enough training data. Conversely, asan emerging technology, Blockchain supports a decentralized architecture. It provides a secure sharingof data and resources to the various nodes of the IoT network is encouraged to remove centralizedcontrol and can overcome the existing challenges in AI. The main goal of our research is to designand develop an IoT architecture with blockchain and AI to support an effective big data analysis. Inthis paper, we propose a Blockchain-enabled Intelligent IoT Architecture with Artificial Intelligencethat provides an efficient way of converging blockchain and AI for IoT with current state-of-the-art techniques and applications. We evaluate the proposed architecture and categorized into twoparts: qualitative analysis and quantitative analysis. In qualitative evaluation, we describe how touse AI and Blockchain in IoT applications with ‘‘AI-driven Blockchain’’ and ‘‘Blockchain-driven AI.’’ Inquantitative analysis, we present a performance evaluation of the BlockIoTIntelligence architecture tocompare existing researches on device, fog, edge and cloud intelligence according to some parameterssuch as accuracy, latency, security and privacy, computational complexity and energy cost in IoT applications.TheevaluationresultsshowthattheproposedarchitectureperformanceovertheexistingIoT architectures and mitigate the current challenges.