Qingji Xue’s research while affiliated with Milwaukee Institute of Art & Design and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (9)


A Distributed VMD-BiLSTM Model for Taxi Demand Forecasting with GPS Sensor Data
  • Article
  • Full-text available

October 2024

·

34 Reads

Sensors

·

Qingji Xue

·

Tianfeng Li

With the ubiquitous deployment of mobile and sensor technologies in modes of transportation, taxis have become a significant component of public transportation. However, vacant taxis represent an important waste of transportation resources. Forecasting taxi demand within a short time achieves a supply–demand balance and reduces oil emissions. Although earlier studies have forwarded highly developed machine learning- and deep learning-based models to forecast taxicab demands, these models often face significant computational expenses and cannot effectively utilize large-scale trajectory sensor data. To address these challenges, in this paper, we propose a hybrid deep learning-based model for taxi demand prediction. In particular, the Variational Mode Decomposition (VMD) algorithm is integrated along with a Bidirectional Long Short-Term Memory (BiLSTM) model to perform the prediction process. The VMD algorithm is applied to decompose time series-aware traffic features into multiple sub-modes of different frequencies. After that, the BiLSTM method is utilized to predict time series data fed with the relevant demand features. To overcome the limitation of high computational expenses, the designed model is performed on the Spark distributed platform. The performance of the proposed model is tested using a real-world dataset, and it surpasses existing state-of-the-art predictive models in terms of accuracy, efficiency, and distributed performance. These findings provide insights for enhancing the efficiency of passenger search and increasing the profit of taxicabs.

Download

FIGURE 2. IND-ID-CCA security model.
FIGURE 3. Setup of Game 1.
FIGURE 3. System model.
FIGURE 4. Setup of Game 1.
FIGURE 5. Challenge of Game 1.

+15

New Constructions of Equality Test Scheme Without Random Oracles

January 2023

·

48 Reads

·

1 Citation

IEEE Access

Huijun Zhu

·

·

Qingji Xue

·

[...]

·

Ao Liu

The proliferation of big data has brought exponential amount of increase in data that is being remotely stored around the globe. Thus, making it imperative to secure the remote data through some encryption mechanism to ensure privacy preservation. However, it often becomes difficult to perform operations over the encrypted data. In order to solve this problem, the equality test function based public key encryption (PKEwET) is proposed. PKEwET approach basically allows secure comparison over encrypted data without revealing the underlying data. This work aims to improve Water’s scheme while introducing a new functionality. More precisly, equality test is being introduced to Water’s scheme so that the encrypted data may be compared without decryption process. To achieve this, an authorization mechanism is being included in which the authorized party uses the trapdoor to test the ciphertext. The scheme is designed under standard model. The security of the proposed scheme is proved with two types of adversaries under the standard model. Finally, the superiority of the proposed scheme in terms of performance is also discussed.


A Hypered Deep-Learning-Based Model of Hyperspectral Images Generation and Classification for Imbalanced Data

December 2022

·

74 Reads

·

6 Citations

Recently, hyperspectral image (HSI) classification has become a hot topic in the geographical images research area. Sufficient samples are required for image classes to properly train classification models. However, a class imbalance problem has emerged in hyperspectral image (HSI) datasets as some classes do not have enough samples for training, and some classes have many samples. Therefore, the performance of classifiers is likely to be biased toward the classes with the largest samples, and this can lead to a decrease in the classification accuracy. Therefore, a new deep-learning-based model is proposed for hyperspectral images generation and classification of imbal-anced data. Firstly, the spectral features are extracted by a 1D convolutional neural network, whereas a 2D convolutional neural network extracts the spatial features and the extracted spatial features and spectral features are catenated into a stacked spatial-spectral feature vector. Secondly, an autoencoder model was developed to generate synthetic images for minority classes, and the image samples were balanced. The GAN model is applied to determine the synthetic images from the real ones and then enhancing the classification performance. Finally, the balanced datasets are fed to a 2D CNN model for performing classification and validating the efficiency of the proposed model. Our model and the state-of-the-art classifiers are evaluated by four open-access HSI datasets. The results showed that the proposed approach can generate better quality samples for rebalancing datasets, which in turn noticeably enhances the classification performance compared to the existing classification models.


Risk Levels Classification of Near-Crashes in Naturalistic Driving Data

May 2022

·

51 Reads

·

10 Citations

Sustainability

Identifying dangerous events from driving behavior data has become a vital challenge in intelligent transportation systems. In this study, we compared machine and deep learning-based methods for classifying the risk levels of near-crashes. A dataset was built for the study by considering variables related to naturalistic driving, temporal data, participants, and road geometry, among others. Hierarchical clustering was applied to categorize the near-crashes into several risk levels based on high-risk driving variables. The adaptive lasso variable model was adopted to reduce factors and select significant driving risk factors. In addition, several machine and deep learning models were used to compare near-crash classification performance by training the models and examining the model with testing data. The results showed that the deep learning models outperformed the machine learning and statistical models in terms of classification performance. The LSTM model achieved the highest performance in terms of all evaluation metrics compared with the state-of-the-art models (accuracy = 96%, recall = 0.93, precision = 0.88, and F1-measure = 0.91). The LSTM model can improve the classification accuracy and prediction of most near-crash events and reduce false near-crash classification. The finding of this study can benefit transportation safety in predicting and classifying driving risk. It can provide useful suggestions for reducing the incidence of critical events and forward road crashes.


Figure 1. System Model.
Figure 2. OW-CCA security model. In Figure 2, O 1 represents the H 1 , H 2 , H 3 , H 4 , and H 5 queries. O 2 (ID) = Extract(msk, ID),
Figure 3. IND-CCA Security Model.
Comparison with other schemes.
Comparison of efficiency with other schemes.
Traceable Scheme of Public Key Encryption with Equality Test

February 2022

·

195 Reads

·

2 Citations

Entropy

Public key encryption supporting equality test (PKEwET) schemes, because of their special function, have good applications in many fields, such as in cloud computing services, blockchain, and the Internet of Things. The original PKEwET has no authorization function. Subsequently, many PKEwET schemes have been proposed with the ability to perform authorization against various application scenarios. However, these schemes are incapable of traceability to the ciphertexts. In this paper, the ability of tracing to the ciphertexts is introduced into a PKEwET scheme. For the ciphertexts, the presented scheme supports not only the equality test, but also has the function of traceability. Meanwhile, the security of the proposed scheme is revealed by a game between an adversary and a simulator, and it achieves a desirable level of security. Depending on the attacker’s privileges, it can resist OW-CCA security against an adversary with a trapdoor, and can resist IND-CCA security against an adversary without a trapdoor. Finally, the performance of the presented scheme is discussed.


Forecasting Taxi Demands Using Generative Adversarial Networks with Multi-Source Data

October 2021

·

215 Reads

·

8 Citations

Applied Sciences

As a popular transportation mode in urban regions, taxis play an essential role in providing comfortable and convenient services for travelers. For the sake of tackling the imbalance between supply and demand, taxi demand forecasting can help drivers plan their routes and reduce waiting time and oil pollution. This paper proposes a deep learning-based model for taxi demand forecasting with multi-source data using Generative Adversarial Networks. Firstly, main features were extracted from multi-source data, including GPS taxi data, road network data, weather data, and points of interest. Secondly, Generative Adversarial Network, comprised of the recurrent network model and the conventional network model, is adopted for fine-grained taxi demand forecasting. A comprehensive experiment is conducted based on a real-world dataset of the city of Wuhan, China. The experimental results showed that our model outperforms state-of-the-art prediction methods and validates the usefulness of our model. This paper provides insights into the temporal, spatial, and external factors in taxi demand-supply equilibrium based on the results. The findings can help policymakers alter the taxi supply and the taxi lease rents for periods and increase taxi profit.


Investigating the Significant Individual Historical Factors of Driving Risk Using Hierarchical Clustering Analysis and Quasi-Poisson Regression Model

April 2020

·

607 Reads

·

15 Citations

Sensors

Driving risk varies substantially according to many factors related to the driven vehicle, environmental conditions, and drivers. This study explores the contributing historical factors of driving risk with hierarchical clustering analysis and the quasi-Poisson regression model. The dataset of the study was collected from two sources: naturalistic driving experiments and self-reports. The drivers who participated in the naturalistic driving experiment were categorized into four risk groups according to their near-crash frequency with the hierarchical clustering method. Moreover, a quasi-Poisson model was used to identify the essential factors of individual driving risk. The findings of this study indicated that historical driving factors have substantial impacts on individual risk of drivers. These factors include the total number of miles driven, the driver's age, the number of illegal parking (past three years), the number of over-speeding (past three years) and passing red lights (past three years). The outcome of the study can help transportation officials, educators, and researchers to consider the influencing factors on individual driving risk and can give insights and provide suggestions to improve driving safety.


Figure 1 Atomic Process in a Petri Net
Figure 2-a Sequential Construction
Figure 4 Part of Flight Purchase Composition Model in OWL-S
Services and their related functions in Flight Ticket System
A Colored Petri Nets Based Model and Verification for Services Composition

October 2019

·

95 Reads

·

2 Citations

Journal of Physics Conference Series

Service composition process generates composite services in order to fullfill service consumer’s requirements that cannot be satisfied by a single service. Literature review addressed services composition but ignored verifying the existence of several serious issues, which may affect Service composition and may lead to failure of the given composite services in the execution time, including consistency of the functionality and QoS Criteria. This paper adopts Colored Petri Nets based model for Services Composition and proposes a QoS aware algorithm for verifying the consistency of composite services. A case study is provided for demonstrating the applicability of the proposed model and algorithm using concepts and values of QoS Criteria of composite services.


Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

August 2018

·

384 Reads

·

28 Citations

Sustainability

With the considerable increase in ownership of motor vehicles, traffic crashes have become a challenge. This paper presents a study of naturalistic driving conducted to collect driving data. The experiments were performed on different road types in the city of Wuhan in China. The collected driving data were used to develop a near-crash database, which covers driving behavior, near-crash factors, driving environment, time, demographics, and experience. A new definition of near-crash events is also proposed. The new definition considers potential risks in driving behavior, such as braking pressure, time headway, and deceleration. A clustering analysis was carried out through a K-means algorithm to classify near-crash events based on their risk level. In addition, a mixed-ordered logit model was used to examine the contributing factors associated with the driving risk of near-crash events. The results indicate that ten factors significantly affect the driving risk of near-crash events: deceleration average, vehicle kinetic energy, near-crash causes, congestion on roads, time of day, driving miles, road types, weekend, age, and experience. The findings may be used by transportation planners to understand the factors that influence driving risk and may provide valuable insights and helpful suggestions for improving transportation rules and reducing traffic collisions thus making roads safer.

Citations (8)


... Very recently, Zhu et al. [5] proposed a new identitybased encryption with equality test (IBEwET) in the standard model, based on Waters' identity-based encryption (IBE) [6]. In [5], the authors claimed that their proposed scheme achieves the indistinguishability against adaptive identity and adaptive chosen ciphertext attacks (IND-ID-CCA) by adversaries who do not have trapdoors for equality tests as well as the one-wayness against adaptive identity and adaptive chosen ciphertext attacks (OW-ID-CCA) by adversaries who have trapdoors. ...

Reference:

Cryptanalysis of Zhu et al.’s Identity-Based Encryption With Equality Test Without Random Oracles
New Constructions of Equality Test Scheme Without Random Oracles

IEEE Access

... Deep learning has also made remarkable achievements in product shape optimization. Naji, H and others realized the intelligent optimization of product shape by learning the shape information in design samples by using deep learning technology [15] . This provides designers with more creative and practical design tools. ...

A Hypered Deep-Learning-Based Model of Hyperspectral Images Generation and Classification for Imbalanced Data

... This highlights that LSTM networks are especially efficient at tasks requiring temporal relationships and intricate sequential patterns. Additionally, Naji et al. [42] found that deep learning models, particularly LSTM networks, outperformed traditional ML models (i.e., SVM, RF and MLP) in classifying the risk levels of near crashes, achieving a classification accuracy of 96%. To effectively apply DL models such as LSTM to datasets such as the one used in this study, an alternative dataset configuration approach would be required to maintain the temporal structure of the driving data recorded at 30-s intervals in order. ...

Risk Levels Classification of Near-Crashes in Naturalistic Driving Data

Sustainability

... These schemes adopt the benefits of identity-based encryption (IBE) cryptosystems, which allow users to generate public and private key pairs using their identity information without the need for digital certificates, therefore addressing the key management problem in public key encryption cryptosystems. This novel approach sparked significant subsequent research on IBEET [17][18][19][20][21][22][23][24][25][26][27][28]. ...

Traceable Scheme of Public Key Encryption with Equality Test

Entropy

... Table 2 describes the attributes of an occupied trip. By isolating occupied trips, the subsequent analysis and prediction models can accurately capture the patterns and trends in the taxi demand of passengers [22]. The visualization in Figure 3 represents taxi trips using small green circles to indicate cruising trips and small blue circles to indicate occupied trips, whereas a large circle represents the event when a passenger is picked up by a taxi, resulting in the taxi's occupied status, or a passenger drop-off from a taxi, wherefore a cruising trip begins. ...

Forecasting Taxi Demands Using Generative Adversarial Networks with Multi-Source Data

Applied Sciences

... importance of various influencing factors and employed an artificial neural network model to classify driver risk under different factor combinations into low-risk, medium-risk, and high-risk drivers [25]. Hasan A.H used hierarchical clustering to classify drivers participating in natural driving experiments into four risk groups based on their near-crash frequencies: conservative, normal, severe, and extremely severe [26]. However, despite these studies, there remains a significant research gap in developing a comprehensive and scientifically rigorous method for quantitatively assessing bus driver risk levels. ...

Investigating the Significant Individual Historical Factors of Driving Risk Using Hierarchical Clustering Analysis and Quasi-Poisson Regression Model

Sensors

... The work [12] considers an approach to automatic composition of services based on the technology of semantic agents. The approach allows you to combine services that are semantically similar and have common goals into a single service. ...

A Colored Petri Nets Based Model and Verification for Services Composition

Journal of Physics Conference Series

... The GROL is an extension of Calfee et al. (2001)'s mixed-effect ROL (MROL) and Fiebig et al. (2010)'s generalized multinomial logit (GMNL). Previous applications using the MROL to analyze BWS data include Fok et al. (2012), Chang et al. (2016), and Naji et al. (2018). Fiebig et al. (2010)'s GMNL has garnered attention in the broader choice modeling literature (e.g., Hole and Kolstad, 2012;Keane and Wasi, 2013;Greene and Hensher, 2010;Balogh et al., 2016;Wright et al., 2018;Liu et al., 2019;Cheng et al., 2021), but the GMNL has been part of an ongoing debate whether scale and preference heterogeneity are separable (Hess and Rose, 2012;Hess and Train, 2017). ...

Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

Sustainability