Completed Projects

27 07, 2021

Support of electronic infrastructure and distance learning requirements due to COVID-19 at the University of West Attica

2023-11-29T11:24:29+03:00

Funding Program: Human Resources Development, Education and Life Lifelong Learning Period: 1/3/2021-31/12/2021 Total Cost: 207.000,00 € MIS: 5094760 Role in the Project: Coordinator.  Dr. Feidakis, Prof. Patrikakis and Prof. Leligkou are part of the team implementing subproject Artificial Intelligence Cloud Infrastructure. Support of electronic infrastructure and distance education due to COVID-19 at the University of West Attica Due to the Covid-19 pandemic, the Greek State imposed distance learning and education. The different course and lab requirements brought the need to use additional special equipment that will facilitate lectures (focusing mainly on theory), in terms of recording specific tutor actions to support theory and lab courses, as well as remote exams.

Support of electronic infrastructure and distance learning requirements due to COVID-19 at the University of West Attica2023-11-29T11:24:29+03:00
5 04, 2021

Βαθιά Μάθηση σε ανεπαρκώς τιτλοφορημένα δεδομένα αισθητήρων για την αναγνώριση χειρονομιών

2022-04-23T00:22:24+03:00

Εφαρμογή τεχνικών Βαθιάς Μάθησης σε ανεπαρκώς τιτλοφορημένα δεδομένα αισθητήρων για την αναγνώριση χειρονομιών - On applying Deep Learning techniques to insufficient labeled sensor data for gesture recognition. Χρηματοδότηση:  € 45.545,50 Επιχειρησιακό Πρόγραμμα «Ανάπτυξη Ανθρώπινου Δυναμικού,Εκπαίδευση και Δια Βίου Μάθηση 2014-2020» Η πράξη συγχρηματοδοτείται από το Ευρωπαϊκό Κοινωνικό Ταμείο (ΕΚΤ). Αντικείμενο της προτεινόμενης έρευνας αποτελεί η ανάπτυξη, δοκιμή και αξιολόγηση ενός μοντέλου Μηχανικής Μάθησης (Machine Learning) το οποίο θα επιτρέπει στους χρήστες φορετών συσκευών (wearable) να κάνουν χρήση προσωποποιημένων εφαρμογών επίγνωσης πλαισίου (context-aware) με στόχο την ευρύτερη βελτίωση της ποιότητας ζωής τους. Κωδικός ΟΠΣ 5050324 Τα αποτελέσματα από το συγκεκριμένο έργο ήταν η συγγραφή δύο επιστημονικών άρθρων τα οποία αποτελούν και παραδοτέα του έργου: «Introducing and benchmarking a one-shot learning gesture recognition dataset» το οποίο υπεβλήθη και παρουσιάστηκε στο διεθνές συνέδριο 10th EAI International Conference on Big Data Technologies and Applications (BDTA) 2020. Το συλλεχθέν σύνολο δεδομένων SENSED (εικόνα 1) παρατίθεται στην πλατφόρμα GitHub για ερευνητικούς σκοπούς. Εικόνα 1: Ενδεικτικοί χαρακτήρες από το σύνολο δεδομένων SENSED «Modality-wise relational reasoning for one-shot sensor-based activity recognition», το οποίο υπεβλήθη στο διεθνές επιστημονικό περιοδικό Elsevier Pattern Recognition Letters και περιλαμβάνεται στη λίστα SCI, με Impact Factor: 3,756. Οι αλγόριθμοι βαθιάς μάθησης (εικόνα 2) που αναπτύχθηκαν παρατίθενται, επίσης, στην πλατφόρμα GitHub για ερευνητικούς σκοπούς. Εικόνα 2: Αλγόριθμος σχεσιακής συλλογιστικής για την αναγνώριση ανθρώπινων δραστηριοτήτων σε ανεπαρκώς τιτλοφορημένα δεδομένα αισθητήρων.

Βαθιά Μάθηση σε ανεπαρκώς τιτλοφορημένα δεδομένα αισθητήρων για την αναγνώριση χειρονομιών2022-04-23T00:22:24+03:00
27 03, 2021

NaLPA-DeeRLe

2022-06-13T12:47:09+03:00

Funding Program: Own funding though CONSERT lab resources Topic: Development and training of Natural Language Processing Agents through the use of Deep Reinforcement Learning Period: 01/02/2021 -  31/01/2022 Total Cost: 6.555 € Role in the Project: Internal CoNSeRT project Description: Τhe objective of this project is to train a Natural Language Processing (NLP) algorithm to generate text based on the collected sparse rewards produced by a Deep Reinforcement Learning (DRL) model. In particular, a Transformer-based Natural Language Generation (NLG) model (e.g.,GPT-2) will be used to create text. At the end of a sentence another Transformer-based model finetuned on a specific task (e.g., RoBERTa on Sentiment Analysis) will evaluate whether the goal has being accomplished (e.g., whether the NLG model has produced a positive comment). Using this pipeline the reward or the penalty of the latter will be backpropagated to the weights of the NLG model based on a DRL algorithm, such as Proximal Policy Optimization (PPO). This approach will be extremely useful for augmenting textual data when it comes to tasks comprised of  few annotated data, or goal-based chatbots that want to accomplish an objective, such as to book a restaurant. Objectives: Goal-based Natural Language Generation Textual data augmentation

NaLPA-DeeRLe2022-06-13T12:47:09+03:00
16 04, 2020

ASSET

2022-04-23T00:12:18+03:00

ASSET - A holistic and Scalable Solution for Research, Innovation and Education in Energy Transition Program/Topic: H2020/LC-SC3-CC-5-2018 - Research, innovation and educational capacities for energy transition Period: 1 May 2019 – 30 April 2021 Total Cost: € 1.999.075 Grant Agreement ID: 837854 Role in the Project: Project leader (Sc. Coordinator Ass. Prof Leligou) Links: https://www.energytransition.academy (Project)   https://cordis.europa.eu/project/id/837854 (Cordis) Description: ASSET´s goal is twofold: first, to create a sustainable and scalable ecosystem including all energy transition and education stakeholders: a) Companies from the energy sector, b) Universities and training actors, c) Authorities and policy makers and d) Society so as to enable a) continuous bottom up creation of research, innovation and educational (RIE) services and b) capacity pooling. By “service” we mean any research, innovation and educational relevant service that can be developed by the educational actors and delivered to the energy sector businesses and society at large. ASSET will also involve actors that can contribute to fuel interdisciplinary research combining SSH disciplines and capable of reaching out to the society to create the new generation of energy-sensitive citizens that are part of the energy transition mission. And second, to deliver the framework and means for continuous collaborative definition of the knowledge-competencies-skills required for the energy transition and for continuous resource pooling to efficiently educate/train large numbers of people in diverse and interdisciplinary topics and carry out research and innovation activities. Participants: Atos Spain S.A., Spain (Coordinator) Università degli Studi di Napoli Federico II, Italy Educational OTE AE, Greece Aalborg Universitet, Denmark Panepistimo Dytikis Attikis, Greece Universitat Politecnica de Valencia, Spain Logical Soft S.r.l., Italy E’Nostra Società Cooperativa, Italy ecopower, Belgium European Association for Storage of Energy, Belgium UNIWA Role: Ass. Prof. Leligou acts as the scientific coordinator of the ASSET project. Prof. Tseles, Ganetsos and Psomopoulos are contributing in learning programmes design and material creation. Prof. Patrikakis and Assistant Professor Karkazis are contributing in the design and creation of the learning graph tool and wide dissemination of the project.

ASSET2022-04-23T00:12:18+03:00
17 10, 2019

Open Courses_1

2022-04-23T00:27:37+03:00

OPEN COURSES Type of Project: Educational / Development Start date: 1/03/2012 End date: 30/11/2015 Description:  Forging Online Education through FIRE (FORGE) is a project proposal bringing the FIRE and eLearning worlds together. FORGE will align FIRE with the ongoing education revolution for mutual benefit. In particular, this project is concerned with specifying development methodologies and best practices for offering FIRE experimentation facilities to learners and to the learning community in general, related both to communications and IT but also to other disciplines including the sciences and social sciences, leading to a strong connection between the learning community and existing FIRE platforms and supporting tools.

Open Courses_12022-04-23T00:27:37+03:00
17 10, 2019

Forge_1

2022-04-23T00:27:32+03:00

FORGE - FORGing online Education through FIRE FIRE (Future Internet Research and Experimentation) Type of Project: FP7 ICT Research project Start date: 1/10/2013 End date: 30/9/2016 Total Cost: 2.413.870 € Role: Affiliate partner Description:  Forging Online Education through FIRE (FORGE) is a project proposal bringing the FIRE and eLearning worlds together. FORGE will align FIRE with the ongoing education revolution for mutual benefit. In particular, this project is concerned with specifying development methodologies and best practices for offering FIRE experimentation facilities to learners and to the learning community in general, related both to communications and IT but also to other disciplines including the sciences and social sciences, leading to a strong connection between the learning community and existing FIRE platforms and supporting tools.

Forge_12022-04-23T00:27:32+03:00
17 10, 2019

Mevlana_1

2022-04-23T00:27:07+03:00

MEVLANA Type of Project: Educational Period: 1/12/2016 - 31/05/2017 Link: http://mevlana.en.ankara.edu.tr/ Description: MEVLANA project supports the exchange of students and professors between Turkish and Greeks Academic Institutions. Piraeus University of Applied Sciences (PUAS) collaborates with Izmir Institute of Technology (IZTECH) for this project for an exchange of tutors (Assistant Pr. Serap Sahin from IZTECH and Associate Pr. Patrikakis Charalampos and Adjunct Pr. Kogias Dimitrios from PUAS) in order to deliver seminar lessons on encryption, privacy, Internet of Things (IoT) and graphs and one student (Burcu Sayin form IZTECH). The student must deliver her thesis during the project, while the results, on the topics that will be covered by the seminar lessons, are encouraged to be submitted as an academic paper in a journal or conference.

Mevlana_12022-04-23T00:27:07+03:00
17 10, 2019

Trillion

2022-04-23T00:26:45+03:00

TRILLION - Trusted, cItizen - Lea coLlaboratIon over sOcial Networks Funding Program: Horizon (H2020-EU.3.7.) Topic: FCT-14-2014 - Ethical/Societal Dimension Topic 2: Enhancing cooperation between law enforcement agencies and citizens - Community policing Type: RIA - Research and Innovation action Period: 1/9/2015 - 31/8/2018 Total Cost: 4.263.408 € Grant agreement: N°653256 Role in the Project:  Scientific Coordinator Link:  https://trillion-project.eng.it/ Cordis link: https://cordis.europa.eu/project/id/653256 Description: TRILLION proposes an open, flexible, secure and resilient socio-technical platform to foster effective collaboration of citizens and law enforcement officers. Using the TRILLION platform, citizens will be able to report crimes, suspicious behavior and incidents, identify hazards and assist law enforcement agents through active participation for achieving better urban security management. On the other hand, Law Enforcement Agencies (LEAs) will be able to detect incidents in a more efficient, content and context aware manner, locate onsite citizens, other LEA representatives and first responders communicate with them, request more information and assign them specific actions to address on-going incidents.

Trillion2022-04-23T00:26:45+03:00
17 10, 2019

Storm_1

2022-04-23T00:25:46+03:00

STORM - Safeguarding cultural heritage through Technical and Organisational Resources Management Funding Program: Horizon (H2020-EU.3.5., H2020-EU.3.7.) Topic: DRS-11-2015 - Disaster Resilience & Climate Change topic 3: Mitigating the impacts of climate change and natural hazards on cultural heritage sites, structures and artefacts Type of Project: Research and Innovation Actions Period: 01/06/2016 - 31/05/2019 Links: http://www.storm-project.eu/ https://cordis.europa.eu/project/rcn/202681/factsheet/en Total Cost: 7.297.875 € Role in the Project:  Scientific Coordinator Description STORM plans to introduce an integrated framework and a platform providing tools and services both at macro level to give a global view of the entire value chain and at specific level to promote the improvement of specific processes for protection and prevention. A novelty of STORM is to promote both views in the same framework; STORM will allow users to address each single issue within a simple process supported by the related technology. The STORM integrated framework will manage those modules to give a view that can be drilled down to give stakeholders the possibility to improve it. To support this, STORM will introduce a system to identify existing processes adding critical relationship management automation to improve the process itself. STORM aims to provide critical decision making tools to all European Cultural Heritage stakeholders affected by climate change and natural hazards. This will be a new innovative capability to improve existing processes related to three identified areas: Prevention, Intervention and Policies, planning and processes. View Demo (EcoBox)

Storm_12022-04-23T00:25:46+03:00
17 10, 2019

DESMOS

2022-10-02T23:33:10+03:00

DESMOS - Smart Interconnected Infrastructures for Security and Protection Funding Program:  Greek National Operational Program “Competitiveness, Entrepreneurship and Innovation – RESEARCH-CREATE-INNOVATE Period: 25/07/2018 - 24/1/2021 Total Cost: 678.239 € Links: https://consert.eee.uniwa.gr/2019/10/17/desmos/ http://desmos-project.gr Role in the Project: Project partner/ WP leader Description: The project aims to develop the DESMOS platform, a framework for the interconnection of smart infrastructures and the use of mobile and computing applications to provide a secure environment for citizens, with emphasis on tourists. The platform will be based on participatory cooperation between people and infrastructure, implementing a framework for the protection of tourists through timely incident reporting, adaptation of interconnected infrastructures in emergency situations, and support for assistance through the actions of volunteers and local authorities. The project sets the following high-level objectives: Fast, timely and accurate notifications in case of emergencies (e.g. allergies, medical incidents, etc.), sending at the same time all contextual information needed to help authorities coordinate and assist people, protecting the privacy of the monitored people. Anonymous reporting of incidents using crowdsourcing, with a special focus on incidents involving tourists, e.g. thefts. Adaptability of services, devices and people to respond to incidents and protect citizens/tourists. In order to realize the abovementioned objectives, the platform will follow a systematic approach for interconnecting people, services and devices using: Applications in mobile and wearable devices that will be used by: Volunteers, who will be timely informed for incidents happening nearby, providing at the same time contextual information about the type of help needed, the current situation and suggestions on how to react. Tourists, who will be able to connect with volunteers and services of the visited area, increasing the feeling of safety and promoting their participation in the safe ecosystem of people and devices. Smart Spots across the area (e.g. municipality) that will be able to listen for reports and requests for help and further propagate them in the local intelligent network, aiming at finding the best solution to mitigate the emergency situation. Fusion and interpretation of heterogeneous events and information through semantic reasoning and decision making for supporting real-time alerts and notification of responsible entities. Researchers in charge of the project: Michalis Feidakis Christos Chatzigeorgiou

DESMOS2022-10-02T23:33:10+03:00
Load More Posts