Welcome to the website of the Robots, Agents, and Interaction (RAI) group at the Department of Data Science & Knowledge Engineering at Maastricht University.
RAI has strong experience in the design and implementation of machine learning, data mining, computer vision, and robotics. This includes broad know-how in research involving learning and coordination among autonomous agents (software or robots), human-robot interaction, human activity & emotion recognition in HCI/HRI (Human-Computer/Robot Interaction), swarm and modular robotics, automated negotiation, and AI knowledge transfer.
Our paper titled “Social Emotion Mining Techniques for Facebook Posts Reaction Prediction” is accepted for publication at the 10th International Conference on Agents and Artificial Intelligence (ICAART 2018) as a full paper.
This paper was authored by four of our Master in Artificial Intelligence students and their supervisor (Florian Krebs, Bruno Lubascher, Tobias Moers, Pieter Schaap, Gerasimos Spanakis) as part of a semester research project and explored how deep learning techniques (CNN and RNN) can be used to set up a prediction module for reaction predictions on Facebook posts.
Our joint work with the Faculty of Health, Medicine and Life Sciences (FHML) ‘A risk score of BMI, HbA1c and triglycerides predicts future glycemic control in type 2 diabetes’ has been accepted for publication at the Diabetes, Obesity and Metabolism Journal.
A description of the platform employed in ICT4Life has been published as a book chapter; it is a joint work of the ICT4Life consortium, in which, RAI, plays a leading role in AAL activities and machine intelligence applications.
For more details:
- A. Sánchez-Rico, P. Garel, I. Notarangelo, M. Quintana, G. Hernández, S. Asteriadis, M. Popa, N. Vretos, V. Solachidis, M. Burgos, A. Girault. ICT Services for Life Improvement for the Elderly. Stud Health Technol Inform. 242:600-605, 2017
ICT4Life project link: http://www.ict4life.eu/
Oour joint work with UPM and CERTH ‘Behaviour analysis through multimodal sensing for improving Parkinson and Alzheimer patients quality of life’ has been accepted for publication at IEEE multimedia magazine
It has been a good week for RAI!
Our paper titled “A retrieval-based dialogue system utilizing utterance and context embeddings” (A. Bartl, G. Spanakis) is accepted as a poster paper at the 16th IEEE International on Machine Learning and Applications (ICMLA2017).
This paper was first-authored by one of our Master in Artificial Intelligence graduates and describes his work towards building a chatbot for one of Maastricht based companies that DKE works with.
Our paper titled “Accumulated Gradient Optimization” (J. Hermans, G. Spanakis, R. Moeckel) is accepted as a full paper at the 9th Asian Conference on Machine Learning (ACML2017).
This paper was first-authored by one of our Master in Artificial Intelligence graduates, based on his thesis work (started at CERN and awarded one of the best thesis awards from DKE for 2017).
Our paper titled “Massive Open Online Courses Temporal Profiling for Dropout Prediction” (T. Rolandus Hagedoorn, G. Spanakis) is accepted as a full paper at the 29th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2017).
This paper was first-authored by one of our Bachelor in Data Science and Knowledge Engineering graduates.
Our paper titled “Reciprocal Recommender System for Users in Massive Open Online Courses (MOOCs)” (S. Prabhakar, G. Spanakis, O. Zaiane) is accepted as a full paper at the 16th International Conference on Web-based Learning (ICWL 2017).
This paper is a joint work with the Computer Science department of University of Alberta.
The University of Maastricht (UM) and Dimokritos (NCSR) are co-organizing the SMAP 2017 Special Session on Multimodal Affective Analysis for Human-Machine Interfaces and Learning Environments (see this link for more information about the SS), which will be part of the SMAP 2017 (smap2017.org) conference in Bratislava, 9th-10th July 2017. The special session will focus on signal analysis and machine learning techniques on affective knowledge from auditory, visual and textual information, employed in human-machine interaction and learning applications and it is organized in collaboration with MaTHiSiS (http://www.mathisis-project.eu/).