BIG-AFF

Fusing multimodal Big Data to provide low-intrusive AFFective and cognitive support in learning contexts

The Spanish Ministry of Education has financed this project with code TIN2014-59641-C2-1-P with 77.319 euro. In this page, you can find more information about the BIG-AFF project.


Objectives

Advance the knowledge on low intrusive approaches for affective data gathering

This project will gather affective data by exploring combinations of new information sources in research experiences and providing new affordable forms of interaction that take the context into account. This approach can also benefit from affect recognition methods, automating the information capture by using low intrusive and low cost infrastructures, as well as knowledge derived from well-known sources in the affect-detection fields.

Explore the application of big data approaches for affect recognition

This project is aimed to expand, investigate, and improve the analysis of the data gathered by using big data techniques to help clarify the fragile nature of that affect detection issue. BIG-AFF project will explore the use of multimodality fusion methods integrated in big data systems to combine different information sources and the management the huge volume of data expected in an e-learning platform following the proposed approach in BIG-AFF. The use of data mining techniques and big data approaches will also be addressed, dealing with massive, heterogeneous and real-time characteristics of multisource unstructured streams of information.

Contribute to key methodological aspects and infrastructure support in emotion modelling (i.e., detection, labelling and support) in educational contexts

In this sense, the BIG-AFF project will explore the design of a new methodological framework to run adequate non-intrusive experiences to identify key emotion modelling aspects (gathering, labelling and supporting) in learning contexts. This methodological framework will build upon conclusions drawn from the results of a series of intra-subject experiences that will guide a large part of our research activities; and will provide the necessary support for the design of personalized, inclusive and adapted affective models that take the user’s characteristics into consideration. We will mainly consider inexpensive and low-intrusive devices that can easily be used in realistic environments. However, we will also look into unrestricted environments, under the assumption that devices that currently are costly and impractical today may become affordable in a near future (e.g. accurate eye tracking devices).

Improve the user’s experience and increase learning gains in educational systems based on previous modelling

Considering the data gathered by applying the methodological framework developed in this project, BIG-AFF will be able to extend previous approaches on both affect-detection and its management. In this context, we understand management as the appropriate usage of educational and psychological criteria in managing affection findings so that personalized actions can be recommended according to the user profile. Here the project will expand methodologies for designing educational recommender systems which are able to combine data mining techniques with educational criteria. The benefits of affective support will be evaluated in different learning contexts. BIG-AFF’s general objetives described above are closely related to the objetives collected in the “Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia, Subprograma Estatal de Generación de Conocimiento, en el marco del Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016” developed under “Orden ECC/1779/2013, de 30 de septiembre”. BIG-AFF is framed in Modality 1 (I+D project), that aims to stimulate the generation of meaningful scientific and technological knowledge by dealing with emergent methodological and technical issues, in our case, related to affective recognition (detection, labelling, processing and support). The research activities proposed in BIG-AFF will improve the validity and reliability of the results obtained in this field, and promote the improvement of the social, technological and educational opportunities of society, specifically those involved in the learning process.



Research Publications

  • JOURNAL PUBLICATIONS
  • Raúl Cabestrero, Pilar Quirós, Olga C. Santos, Sergio Salmeron-Majadas, Raul Uria-Rivas, Jesús González-Boticario, David Arnau, Miguel Arevalillo-Herráez, Francesc J. Ferri: Some insights into the impact of affective information when delivering feedback to students. Behaviour & IT 37(12): 1252-1263 (2018)
  • Aladdin Ayesh, Miguel Arevalillo-Herráez, Pablo Arnau-González: SOM-Based Class Discovery for Emotion Detection Based on DEAP Dataset. International Journal of Software Science and Computational Intelligence 10(1): 15-26 (2018)
  • Cristina Cunha-Perez, Miguel Arevalillo-Herráez, Luis Marco-Giménez, David Arnau: On Incorporating Affective Support to an Intelligent Tutoring System: an Empirical Study. IEEE-RITA 13(2): 63-69 (2018)
  • Miguel Arevalillo-Herráez, Luis Marco-Giménez, David Arnau, José Antonio González-Calero: Adding sensor-free intention-based affective support to an Intelligent Tutoring System. Knowledge-Based Systems 132: 85-93 (2017)
  • Pablo Arnau-González, Miguel Arevalillo-Herráez, Naeem Ramzan: Fusing highly dimensional energy and connectivity features to identify affective states from EEG signals. Neurocomputing 244: 81-89 (2017)
  • José Antonio González-Calero, David Arnau, Luis Puig, Miguel Arevalillo-Herráez: Intensive scaffolding in an Intelligent Tutoring System for the learning of algebraic word problem solving. British Journal in Educational Technology 46(6): 1189-1200 (2015)
  • Salvador Moreno-Picot, Francesc J. Ferri, Miguel Arevalillo-Herráez, Wladimiro Díaz Villanueva: Efficient Analysis and Synthesis Using a New Factorization of the Gabor Frame Matrix. IEEE Transactions on Signal Processing 66(17): 4564-4573 (2018)
  • Emilia López-Iñesta, Francisco Grimaldo, Miguel Arevalillo-Herráez: Combining feature extraction and expansion to improve classification based similarity learning. Pattern Recognition Letters 93: 95-103 (2017)
  • Emilia López-Iñesta, Francisco Grimaldo, Miguel Arevalillo-Herráez: Learning Similarity Scores by using a Family of Distance Funcions in Multiple feature Spaces. International Journal of Pattern Recognition and Artificial Intelligence 31(8): 1-21 (2017)
  • Aladdin Ayesh, Miguel Arevalillo-Herráez, Francesc J. Ferri, Towards Psychologically based Personalised Modelling of Emotions Using Associative Classifiers. International Journal of Cognitive Informatics and Natural Intelligence, Vol. 10, No. 2, pp 52-64, 201
  • CONFERENCE PUBLICATIONS
  • Aladdin Ayesh, Miguel Arevalillo-Herráez, Pablo Arnau-González: Class discovery from semi-structured EEG data for affective computing and personalisation. 16th IEEE ICCI*CC 2017: Oxford, UK: 96-101
  • Pablo Arnau-González, Stamos Katsigiannis, Naeem Ramzan, Debbie Tolson, Miguel Arevalillo-Herráez: ES1D: A Deep Network for EEG-Based Subject Identification. 17th IEEE International Conference on Bioinformatics and Bioengineering, BIBE 2017, Washington, DC, USA, October 23-25, 2017: 81-85
  • María T. Sanz, David Arnau, José Antonio González-Calero, Francesc J. Ferri, Miguel Arevalillo-Herráez: Predicting human performance in interactive tasks by using dynamic models. 2017 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2017, Banff, AB, Canada, October 5-8: 776-780
  • Miguel Arevalillo-Herráez, David Arnau, Francesc J. Ferri, Olga C. Santos: GUI-driven intelligent tutoring system with affective support to help learning the algebraic method. 2017 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2017, Banff, AB, Canada, October 5-8: 2867-2872
  • Jesus Boticario, Olga C. Santos, Raúl Cabestrero, Pilar Quirós, Sergio Salmeron-Majadas, Raul Uria-Rivas, Mar Saneiro, Miguel Arevalillo-Herráez, Francesc J. Ferri: BIG-AFF: Exploring Low Cost and Low Intrusive Infrastructures for Affective Computing in Secondary Schools. Adjunct Publication of the 25th Conference on User Modeling, Adaptation and Personalization, UMAP 2017, Bratislava, Slovakia, July 09 - 12, 2017: 287-292
  • Miguel Arevalillo-Herráez, Aladdin Ayesh, Olga C. Santos, Pablo Arnau-González: Combining Supervised and Unsupervised Learning to Discover Emotional Classes. UMAP 2017: Bratislava, Slovakia: 355-356
  • María T. Sanz, David Arnau, José Antonio González-Calero, Miguel Arevalillo-Herráez: Using System Dynamics to Model Student Performance in an Intelligent Tutoring System. UMAP 2017: Bratislava, Slovakia : 385-386
  • Luis Marco-Giménez, Miguel Arevalillo-Herráez, Francesc J. Ferri, Salvador Moreno-Picot, Jesus Boticario, Olga C. Santos, Sergio Salmeron-Majadas, Mar Saneiro, Raul Uria-Rivas, David Arnau, José Antonio González-Calero, Aladdin Ayesh, Raúl Cabestrero, Pilar Quirós, Pablo Arnau-González, Naeem Ramzan: Affective and Behavioral Assessment for Adaptive Intelligent Tutoring Systems. Late-breaking Results, Posters, Demos, Doctoral Consortium and Workshops Proceedings of the 24th ACM Conference on User Modeling, Adaptation and Personalisation (UMAP 2016), Halifax, Canada, July 13-16, 2016
  • Pablo Arnau-González, Naeem Ramzan, Miguel Arevalillo-Herráez, A method to identify affect levels from EEG signals using two dimensional emotional levels, 30th European Simulation and Modelling Conference (ESM'2016), October 26-28, 2016, Las Palmas de Gran Canaria, Spain
  • Luis Marco-Giménez, Miguel Arevalillo-Herráez, Aladdin Ayesh and Mariusz Szwoch, An Extension to the FEEDB Multimodal Database of Facial Expressions and Emotions, 29th European Simulation and Modelling Conference (ESM'2015): 455-460, October 26-28, 2015, Holiday Inn, Leicester, UK