Choquet integral logistic regression algorithm based on L-mesure and γ-support

• Published in: 2008 International Conference on Wavelet Analysis and Pattern Recognition Vol. 2; pp. 771 - 776
• Main Authors: Hsiang-Chuang Liu, Yu-Du Jheng, Guey-Shya Chen, Bai-Cheng Jeng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2008
• Subjects: Artificial neural networks; Classification algorithms; Equations; Fuzzy measure; hoquet integral; L-measure; Logistics; Mathematical model; Support vector machines; Wavelet analysis; γ-support; λ-measure
• ISBN: 9781424422388; 1424422388
• ISSN: 2158-5695
• DOI: 10.1109/ICWAPR.2008.4635881

Logistic regression algorithm and SVM algorithm are two well-known classification algorithms but when the multi-collinearity between independent variables occurs in above two algorithms, their classifying performance will always be not good. An improved classification algorithm combining the Choquet integral with respect to the lambda-measure based on gamma-support is proposed by our previous work. In this paper, we replaced the more sensitive fuzzy measure, L-measure with the lambda-measure in above improved classification algorithm, and we obtained a further improved algorithm, called Choquet integral logistic regression algorithm based on L-measure and gamma-support. For evaluating the performances of the SVM, logistic regression and the Choquet integral logistic regression algorithm with gamma-support based on P-measure, lambda-measure and L-measure, respectively, a real data experiment by using leave-one-out cross-validation accuracy is conducted. Experimental result shows that our new algorithm has the best performance.