Библиографическое описание:Classification Algorithm with Lexicase Selection : доклад, тезисы доклада / T. T. Pleshkova, V. V. Stanovov. - [S. l. : s. n.], 2022. - Текст : непосредственный // HYBRID METHODS OF MODELING AND OPTIMIZATION IN COMPLEX SYSTEMS : Proceedings of the International Workshop “Hybrid methods of modeling and optimization in complex systems” (HMMOCS 2022) / Krasnoyarsk Regional Science and Technology City Hall, Russia Siberian Federal University, Reshetnev Siberian State University of Science and Technology ; International Workshop “Hybrid methods of modeling and optimization in complex systems” (HMMOCS 2022) (2022 ; 22.11 - 24.11 ; Krasnoyarsk). - London, United Kingdom : European Proceedings, 2022. - P. 194-200. - ISBN 9781802969603, DOI 10.15405/epct.23021.24.
Аннотация:The hybrid fuzzy genetic based algorithm was implemented. The genetic fuzzy systems are applied for classifier learning, and although they allow creating interpretable rule bases, the process of designing a rule base can be improved with specific genetic operators, such as lexicase selection. The influence of lexicase selection on the efficiency of its work for data classification is examined in this article. “Ring”, “Phoneme” and “Satimage” datasets were used for verification. The results were analyzed by verification using Mann-Whitney U test. According to the results of basic hybrid fuzzy genetic based algorithm with lexicase selection of several runs, where in each the duration of the search for the best rule base was limited to five hundred generations, efficiency was only on number of rules, but paired with the previously developed initialization modification, the accuracy and F-score were improved on “Satimage” dataset and the number of rules decreased on all of the datasets.
