Title :
A Fusion of Multiple Classifiers Approach Based on Reliability function for Text Categorization
Author :
Chen, Qingxuan ; Zheng, Dequan ; Zhao, Tiejun ; Li, Sheng
Author_Institution :
MOE-MS Key Lab. of Natural Language Process. & Speech, Harbin Inst. of Technol., Harbin
Abstract :
With the development of Internet and the rapid expansion of electronic resource, text classification technology is becoming an effective organization and management tool to deal with information. In this paper, a method for text categorization based on the fusion of multiple classifiers was presented, reliability function was introduction to select the text that hard to give determine by the main classifier, for these texts, multiple classifiers were used to give the determine which category the unlabeled documents belong to by voting. Experiments showed that the performance of text classification improved by the proposed method. Compared with single classifier, this method achieved better performance, only increasing a small amount of time than using single main classifier. Besides this, this method is more stable than using single classifier for text categorization task, especially when using different corpuses to check the performance of various methods.
Keywords :
Internet; classification; text analysis; Internet; multiple classifiers; reliability function; text categorization; Frequency shift keying; Fuzzy systems; Internet; Laboratories; Natural language processing; Speech processing; Support vector machine classification; Support vector machines; Text categorization; Voting; main classifier; multiple classifiers; reliability function; text categorization;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery, 2008. FSKD '08. Fifth International Conference on
Conference_Location :
Shandong
Print_ISBN :
978-0-7695-3305-6
DOI :
10.1109/FSKD.2008.373