Panagiotis KasnesisUniversity of West Attica | TEIATH · Department of Electrical and Electronics Engineering
Panagiotis Kasnesis
Doctor of Engineering
About
40
Publications
6,310
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
276
Citations
Citations since 2017
Introduction
Publications
Publications (40)
![Fig. 2. Illustrations of the 5 selected Emergency Arm Signals [24].](publication/364443006/figure/fig2/AS:11431281091064941@1666286880482/Illustrations-of-the-5-selected-Emergency-Arm-Signals-24_Q320.jpg)
Efficient and quick remote communication in search and rescue operations can be life-saving for the first responders. However, while operating on the field means of communication based on text, image and audio are not suitable for several disaster scenarios. In this paper, we present a smartwatch-based application, which utilizes a Deep Learning (D...
Nowadays, Hearth Rate (HR) monitoring is a key feature of almost all wrist-worn devices exploiting photoplethysmography (PPG) sensors. However, arm movements affect the performance of PPG-based HR tracking. This issue is usually addressed by fusing the PPG signal with data produced by inertial measurement units. Thus, deep learning algorithms have...
Patents are documents that contain state of the art technical and scientific information in almost every field of science. Patent applications filed every day in the Hellenic Industrial Property Organization, have to be intellectually classified by domain experts based on a hierarchical taxonomy. As a result, this classification process is labor in...
Graffiti is common in many communities and even affects our historical and heritage structures. This leads to a decrease in the revenue associated with commercial activities or services (e.g., shops, restaurants, residences), and potentially reduces tourism in a region. Visual data, in the form of photographs, is becoming an efficient mechanism to...
Search and Rescue (SaR) dogs are important assets in the hands of first responders, as they have the ability to locate the victim even in cases where the vision and or the sound is limited, due to their inherent talents in olfactory and auditory senses. In this work, we propose a deep-learning-assisted implementation incorporating a wearable device...
The widespread use of social networks has brought to the foreground a very important issue, the veracity of the information circulating within them. Many natural language processing methods have been proposed in the past to assess a post’s content with respect to its reliability; however, end-to-end approaches are not comparable in ability to human...
A widespread practice in machine learning solutions is the continuous use of human intelligence to increase their quality and efficiency. A common problem in such solutions is the requirement of a large amount of labeled data. In this paper, we present a practical implementation of the human-in-the-loop computing practice, which includes the combin...
We show that the combination of machine learning methods and analytical expressions can enhance the OSNIR estimation of an optical link of up to 2 dB compared with the use of an analytical expression alone. For this purpose, we exploit seven machine learning algorithms and we examine their OSNIR improvement for 3,000 different operational cases.
Identifying the origin of information posted on social media and how this may have changed over time can be very helpful to users in determining whether they trust it or not. This currently requires disproportionate effort for the average social media user, who instead has to rely on fact-checkers or other intermediaries to identify information pro...
Deep learning techniques have been widely and successfully applied, over the last five years, to recognize the gestures and activities performed by users wearing electronic devices. However, the collected datasets are built in an old fashioned way, mostly comprised of subjects that perform many times few different gestures/activities. This paper ad...
Deep learning concepts have been successfully transferred from the computer vision task to that of wearable human activity recognition (HAR) over the last few years. However, deep learning models require a large volume of annotated samples to be efficiently trained, while adding new activities results in training the whole network from scratch. In...
Deep learning techniques have been widely and successfully applied, over the last five years, to recognize the gestures and activities performed by users wearing electronic devices. However, the collected datasets are built in an old fashioned way, mostly comprised of subjects that perform many times few different gestures/activities. This paper ad...
Monitoring the behavior of animals (e.g., eating habits) can lead to conclusions regarding animal's welfare. To achieve this, remote monitoring of animal activity with the aid of inertial sensors and use of machine learning algorithms over the collected data can be used. However, these algorithms rely on handcrafted features extracted by statistica...
Identifying the provenance of information posted on social media and how this information may have changed over time can be very helpful in assessing its trustworthiness. Here, we introduce a novel mechanism for discovering "post-based" information cascades, including the earliest relevant post and how its information has evolved over subsequent po...
Using the “technology as the solution” line of thought but with the added twist of putting a human in the loop, a team of eight coauthors led by Greek academic Panagiotis Monachelis proposes to combine peer-to-peer decentralized networks and blockchain technology to address the challenge of misinformation in social media. The authors provide a deta...
Users of traditional centralised social media networks have limited knowledge about the original source of information and even less about its trustworthiness and how this information has spread and been modified. Existing media verification tools include websites or browser add-ons that are closed-source or centralised, or they do not include user...
We present the technologies and the theoretical background of an intelligent interconnected infrastructure for public security and safety. The innovation of the framework lies in the intelligent combination of devices and human information towards human and situational awareness, so as to provide a protection and security environment for citizens....
Taking advantage of the processing power that the modern smartphones possess, we advocate a privacy aware approach to predict and suggest to travelers' places of interest. This approach solves a significant privacy flaw that exists in prevalent tourism applications.
Modern e-learning and distance learning systems suffer severe lack of affect-aware interaction: the typical system is irresponsive to the affective state of the user, while even an inadequate human tutor would respond to it and even adapt his/her instruction accordingly. The main goal of this paper is to describe a scenario that deploys state-of-th...
Despite the visionary tenders that emerging technologies bring to education, modern learning environments such as MOOCs or Webinars still suffer from adequate affective awareness and effective feedback mechanisms, often leading to low engagement or abandonment. Artificial Conversational Agents hold the premises to ease the modern learner’s isolatio...
Cultural heritage sites, apart from being the tangible link to a country's history and culture, actively contribute to the national economy, offering a foundation upon which cultural tourism can develop. This importance at the cultural and economic level, advocates for the need for preservation of cultural heritage sites for the future generations....
Human Activity Recognition (HAR) based on motion sensors has drawn a lot of attention over the last few years, since perceiving the human status enables context-aware applications to adapt their services on users’ needs. However, motion sensor fusion and feature extraction have not reached their full potentials, remaining still an open issue. In th...
Climatic changes and intensive industrialization have contributed to increasing the risk of damage of cultural heritage (CH) artefacts. On the other hand, small and medium-sized museums, as well as small CH sites struggle to fulfil international recommendations for protection and conservation, due to budget limitations. The constantly increasing po...
It has been decades since the research community has put efforts into combining human and machine intelligence. With the rapidly surging of mobile sensing, gamification techniques have contributed on making the crowd-sourcing computing techniques a promising paradigm for large-scale data sensing. In this paper, starting from a study of gamification...
Human Activity Recognition (HAR) based on motion sensors has drawn a lot of attention over the last few years, since perceiving the human status enables context-aware applications to adapt their services on users' needs. However, motion sensor fusion and feature extraction have not reached their full potentials, remaining still an open issue. In th...
Situational and context awareness is becoming more and more important on the course towards intelligent machines and devices, offering a comprehensive toolset for improving our quality of life. The increased computational capacity of personal/smart devices, and their constantly increasing capabilities for sensing, allow for a large amount of collec...
Fog Computing is an emerging paradigm, suitable to serve the particular needs of IoT networks. It includes the deployment of computational devices at the edge of the network facilitating faster real-time processing of time-sensitive data. In this article, we present a Fog architecture, which diverges from the traditional hierarchical and centralize...
The advent of the Internet of Everything, where things and data providers can connect not only to other things and data providers, but to human entities as well, and are enriched with intelligence, calls for sophisticated data handling, storing, and sharing mechanisms. In this chapter, we present an ecosystem built over the idea of Internet of Ever...
Situational and context awareness are becoming more and more important on the road toward intelligent machines and devices that can offer a comprehensive toolset for improving quality of life. The increased computational capacity of personal and smart devices, and their constantly increasing capabilities for sensing, allow a large amount of collect...
The concept of an Internet of Things is already mature, enough to start evolving towards an Internet of Everything, over which humans, objects and virtual items can be interconnected. For this to be feasible though, there is the need for a framework which not only supports the interconnection between the entities of this Internet, but also allows t...
The rapidly increasing number of health monitoring devices, mainly wearables, will let users monitor their health and habitual parameters in a simple and easy way. However, the diversity of available platforms calls for simplified, standard solutions that can offer integrated tools to users -that is, patients, professionals, and practitioners. The...
The enormous set of health and wellbeing data sources, as well as the diversity of the data, calls for an effective, time-aware integration paradigm that aids at the manipulation of the information by experts as a whole and not as individual pieces of knowledge. In this paper, we present the Health Service Bus, a service-based platform built on top...
Most scientific applications tend to have a very resource demanding nature and the simulation of such scientific problems often requires a prohibitive amount of time to complete. Distributed computing offers a solution by segmenting the application into smaller processes and allocating them to a cluster of workers. This model was widely followed by...
Projects
Projects (3)
Goals:
- Built and benchmark a one-shot wearable-based gesture recognition dataset.
- Examine deep learning model architectures applied to one-shot sensor-based HAR.
- Introduce a modality-wise relational network to discover activity similarities.
- Extensively investigate multi-head self-attention performance on one-shot HAR.
- Construct explainable attention maps to display the modality-wise similarities.
This research is co-financed by Greece and the European Union (European Social Fund- ESF) through the Operational Programme <<Human Resources Development, Education and Lifelong Learning 2014-2020>> in the context of the project “On applying Deep Learning techniques to insufficient labeled sensor data for gesture recognition” (MIS 5050324).
EUNOMIA uses a positive-first approach to the information trustworthiness challenge in social media. It will be a fully decentralised peer-to-peer platform where a blockchain infrastructure will allow a participating user to easily show that they are the original source of a piece of information posted online, whether this is an important news item or just a joke. Also, when a EUNOMIA user sees a post online, they will will be able to show not only where information started from, but also if and how it has been modified in an information cascade visualised at the press of a button.
As different users have different ways of determining themselves whether a piece of information is indeed trustworthy or not from their point of view, EUNOMIA will provide the information of their preference in the way they decide. This may include information that indicates bot activity, such as the ratio of followers to following, and other indicators suggested by the users themselves or identified in the scientific literature.
Citizen participation is actively encouraged in content verification by voting on content trustworthiness. The aim is that the users will take ownership of the problem of disinformation, not to rely on third party fact-checkers or computer software to do it for them. The number of votes will appear as one of the several indicators that the user will have easy access to if they wish to visualise along the information cascade of the post they are viewing.
EUNOMIA will create a social media companion in both mobile and desktop versions, that will visualise on the information cascade the information trustworthiness indicators chosen by the user. EUNOMIA will be tested on new forms of decentralised media because of the privacy-first outlook of their communities, their open-source and transparent nature, and the technological challenge that they entail. EUNOMIA's technologies will be tested in specifically created new instances of Mastodon and Diaspora with users participating for the experimental evaluation.
WHO IS BEHIND THE PROJECT?
A partnership of academic, non-private, and state organizations and private business aiming to interrupt the cycle of fake news. The project is funded by a 2.9 million Euro grant awarded by the highly competitive EU Horizon 2020 programme. To achieve this goal, a highly complementary consortium of 10 partners from 9 countries has joined forces with an advisory board of experts from around the world.
Website: https://eunomia.social
Twitter: https://twitter.com/projectEUNOMIA
Mastodon: https://mastodon.social/@Eunomia
Diaspora: https://diasp.org/people/d7d9e28028bb0138b22f047d7b62795e
