Learning General Preference Models from Physiological Responses in Video Games: How Complex Is It?
An affective preference model can be successfully learnt from pairwise comparison of physiological responses. Several approaches to do this obtain different performances. The higher ranked seem to use non linear models and complex feature selection strategies. We present a comparison of three linear and non linear classification methods, combined with a simple and a complex feature selection strategy (sequential forward selection and a genetic algorithm), on two datasets. We apply a strict crossvalidation framework to test the generalization capability of the models when facing physiological data coming from a new user. We show that, when generalization is the goal, complex non-linear models trained using fancy strategies might easily get trapped by overfitting, while linear ones might be preferable. Although this could be expected, the only way to appreciate it has to pass through proper cross-validation, and this is often forgot when rushing in the “best” performance challenge.
Keywordsenjoyment evaluation physiological signals feature selection SFS neuro evolution emotion in games cross validation
Unable to display preview. Download preview PDF.
- 1.Broek, E.L.V.D., Janssen, J.H., Westerink, J.: Guidelines for affective signal processing (asp): From lab to life, pp. 1–6. IEEE, Los Alamitos (2009)Google Scholar
- 2.Kolodyazhniy, V., Kreibig, S., Gross, J., Roth, W., Wilhelm, F.: An affective computing approach to physiological emotion specificity: Toward subject-independent and stimulus-independent classification of film-induced emotions. Psychophysiology (2011)Google Scholar
- 5.Martínez, H.P., Yannakakis, G.N.: Genetic search feature selection for affective modeling: a case study on reported preferences. In: Proceedings of the 3rd International Workshop on Affective Interaction in Natural Environments, AFFINE 2010, pp. 15–20. ACM, New York (2010)Google Scholar
- 7.Tognetti, S., Garbarino, M., Bonarini, A., Matteucci, M.: Enjoyment recognition from physiological data in a car racing game. In: Proceedings of the 3rd International Workshop on Affective Interaction in Natural Environments, AFFINE 2010, pp. 3–8. ACM, New York (2010)Google Scholar
- 9.Yannakakis, G.: Preference learning for affective modeling. In: Proceeding of the International Conference on Affective Computing and Intelligent Interaction, ACII 2009, pp. 1–6. IEEE, Los Alamitos (2009)Google Scholar
- 11.Baker, J.E.: Reducing bias and inefficiency in the selection algorithm, pp. 14–21. Lawrence Erlbaum Associates, Mahwah (1987)Google Scholar