DocumentCode :
3694319
Title :
Research on clothing product reviews mining based on the maximum entropy
Author :
Qinghong Yang; Pengfei Feng
Author_Institution :
School of Software, Beihang University, Xueyuan Road, Haidian District, Beijing, 100191, China
fYear :
2015
Firstpage :
255
Lastpage :
258
Abstract :
This paper excavated the review theme of clothing products by method of association rules, and built a maximum entropy model for the reviews classification. Then this paper did experimental verification to large-scale clothing product reviews classification, which verified the practical effect that maximum entropy model had in the comment text classification problems. In the process of classification, the maximum entropy model had a good effect, of which accuracy was over 90%.
Keywords :
"Entropy","Optimization","Accuracy","Computational modeling"
Publisher :
ieee
Conference_Titel :
Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE), 2015 11th International Conference on
Type :
conf
Filename :
7332578
Link To Document :
بازگشت