Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs
Training neural networks in distinguishing different emotions from physiological signals frequently involves fuzzy definitions of each affective state. In addition, manual design of classification tasks often uses sub-optimum classifier parameter settings, leading to average classification performan...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/13925/1/Affect_classification_using_genetic-optimized_ensembles_of.pdf |
| _version_ | 1825720439081533440 |
|---|---|
| author | Liew, W.S. Seera, M. Loo, C.K. Lim, E. |
| author_facet | Liew, W.S. Seera, M. Loo, C.K. Lim, E. |
| author_sort | Liew, W.S. |
| collection | UM |
| description | Training neural networks in distinguishing different emotions from physiological signals frequently involves fuzzy definitions of each affective state. In addition, manual design of classification tasks often uses sub-optimum classifier parameter settings, leading to average classification performance. In this study, an attempt to create a framework for multi-layered optimization of an ensemble of classifiers to maximize the system's ability to learn and classify affect, and to minimize human involvement in setting optimum parameters for the classification system is proposed. Using fuzzy adaptive resonance theory mapping (ARTMAP) as the classifier template, genetic algorithms (GAs) were employed to perform exhaustive search for the best combination of parameter settings for individual classifier performance. Speciation was implemented using subset selection of classification data attributes, as well as using an island model genetic algorithms method. Subsequently, the generated population of optimum classifier configurations was used as candidates to form an ensemble of classifiers. Another set of GAs were used to search for the combination of classifiers that would result in the best classification ensemble accuracy. The proposed methodology was tested using two affective data sets and was able to produce relatively small ensembles of fuzzy ARTMAPs with excellent affect recognition accuracy. (C) 2014 Elsevier B.V. All rights reserved. |
| first_indexed | 2024-03-06T05:35:09Z |
| format | Article |
| id | um.eprints-13925 |
| institution | Universiti Malaya |
| language | English |
| last_indexed | 2024-03-06T05:35:09Z |
| publishDate | 2015 |
| record_format | dspace |
| spelling | um.eprints-139252015-09-20T23:58:38Z http://eprints.um.edu.my/13925/ Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs Liew, W.S. Seera, M. Loo, C.K. Lim, E. T Technology (General) TA Engineering (General). Civil engineering (General) Training neural networks in distinguishing different emotions from physiological signals frequently involves fuzzy definitions of each affective state. In addition, manual design of classification tasks often uses sub-optimum classifier parameter settings, leading to average classification performance. In this study, an attempt to create a framework for multi-layered optimization of an ensemble of classifiers to maximize the system's ability to learn and classify affect, and to minimize human involvement in setting optimum parameters for the classification system is proposed. Using fuzzy adaptive resonance theory mapping (ARTMAP) as the classifier template, genetic algorithms (GAs) were employed to perform exhaustive search for the best combination of parameter settings for individual classifier performance. Speciation was implemented using subset selection of classification data attributes, as well as using an island model genetic algorithms method. Subsequently, the generated population of optimum classifier configurations was used as candidates to form an ensemble of classifiers. Another set of GAs were used to search for the combination of classifiers that would result in the best classification ensemble accuracy. The proposed methodology was tested using two affective data sets and was able to produce relatively small ensembles of fuzzy ARTMAPs with excellent affect recognition accuracy. (C) 2014 Elsevier B.V. All rights reserved. 2015-02 Article PeerReviewed application/pdf en http://eprints.um.edu.my/13925/1/Affect_classification_using_genetic-optimized_ensembles_of.pdf Liew, W.S. and Seera, M. and Loo, C.K. and Lim, E. (2015) Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs. Applied Soft Computing, 27. pp. 53-63. ISSN 1568-4946, DOI https://doi.org/10.1016/j.asoc.2014.10.032 <https://doi.org/10.1016/j.asoc.2014.10.032>. http://www.sciencedirect.com/science/article/pii/S1568494614005444 DOI 10.1016/j.asoc.2014.10.032 |
| spellingShingle | T Technology (General) TA Engineering (General). Civil engineering (General) Liew, W.S. Seera, M. Loo, C.K. Lim, E. Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title | Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title_full | Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title_fullStr | Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title_full_unstemmed | Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title_short | Affect classification using genetic-optimized ensembles of fuzzy ARTMAPs |
| title_sort | affect classification using genetic optimized ensembles of fuzzy artmaps |
| topic | T Technology (General) TA Engineering (General). Civil engineering (General) |
| url | http://eprints.um.edu.my/13925/1/Affect_classification_using_genetic-optimized_ensembles_of.pdf |
| work_keys_str_mv | AT liewws affectclassificationusinggeneticoptimizedensemblesoffuzzyartmaps AT seeram affectclassificationusinggeneticoptimizedensemblesoffuzzyartmaps AT loock affectclassificationusinggeneticoptimizedensemblesoffuzzyartmaps AT lime affectclassificationusinggeneticoptimizedensemblesoffuzzyartmaps |