Title :
Generalization Ability in SVM with Fuzzy Class Labels
Author_Institution :
Dept. of Mech. Eng., Northern Taiwan Inst. of Sci. & Technol., Taipei
Abstract :
The paper attempts to introduce a fundamental fuzzy concept to break the equivalent attitude of the input training set of SVM, and tries to give individual example in the set a different attitude. The attitude can stand for the influence that the example takes into account in the classification. In the paper, we present a method to refresh the attitude by assigning proper fuzzy value to the class label of each example. Based on the benefit of individualized fuzzification, we re-examine the formulation of SVM, and try to find out the corresponding effects. Although it costs a little more computation, the introduction of fuzzy class label is eventually worth for the better generalization performance with a large margin. The effect is especially both crucial and subtle in worse confused training data set with many hard examples
Keywords :
fuzzy set theory; generalisation (artificial intelligence); support vector machines; fuzzification; fuzzy class label; generalization ability; support vector machine; Costs; Error correction; Fuzzy sets; Paper technology; Statistical learning; Statistics; Strips; Support vector machine classification; Support vector machines; Training data;
Conference_Titel :
Computational Intelligence and Security, 2006 International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
1-4244-0605-6
Electronic_ISBN :
1-4244-0605-6
DOI :
10.1109/ICCIAS.2006.294097