An IG-RS-SVM classifier for analyzing reviews of E-commerce product
Jiajun Ye, Huan Ren, Hangxia Zhou · 2015
Analyzing reviews of E-commerce product is a kind of text classification which belongs to supervised learning.Due to the huge number of words, high dimensional feature space is a serious problem in text classification.In order to solve it, a new algorithm, IG-RS-SVM, is proposed.Information Gain (IG) is a feature selection algorithm which can reduce the dimension of feature subspace.Random subspace, a kind of ensemble learning algorithm, can divide the feature space to smaller ones each submitted to a base classifier such as Support Vector Machine (SVM).After experiments, it shows that IG-RS-SVM algorithm can effectively improve the text classification accuracy.