DocumentCode :
3431438
Title :
Research on grain information classification based on SVM decision tree
Author :
Geng, Ruihuan ; Zhang, Dexian ; Chai, Jiajia
Author_Institution :
College of Information Science and Engineering, Henan University of Technology, Zhengzhou, China
fYear :
2012
fDate :
11-13 Aug. 2012
Firstpage :
138
Lastpage :
141
Abstract :
The defections of traditional support vector machine (for short SVM) are analyzed in the paper. According to the characteristics of grain information on the web, a multi-class classification method based on SVM decision tree (for short SVM-DT) is presented for grain information classification. Experiments prove that F1-Measure values for SVM-DT algorithm is superior to the traditional SVM algorithm. It is more suitable for application to grain information classification system.
Keywords :
Frequency measurement; Noise; Support vector machines; Grain information; SVM Binary tree; Web text mining;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Granular Computing (GrC), 2012 IEEE International Conference on
Conference_Location :
Hangzhou, China
Print_ISBN :
978-1-4673-2310-9
Type :
conf
DOI :
10.1109/GrC.2012.6468622
Filename :
6468622
Link To Document :
بازگشت