Research on Chinese Sentiment Recognition Using Dynamic Feature Selection Method
Liu Zh · Journal of Chinese Computer Systems · 2014
Due to the high sparsity and redundancy of feature space in Chinese sentiment recognition research,a dynamic feature selection mechanism based sentiment recognition method is proposed in terms of ensemble learning.This method first constructs the distribution of feature subset dimensionality and importance distribution of feature space via kernel smoothing method.Then the whole feature space is adaptively divided into multiple subspaces with different granularity,and a base classifier is built on corresponding subspace.Finally,a majority voting method is employed to fuse these base classifiers to form an ensemble recognition model.The experiments are conducted on the review posts collected from the campus BBS.The results show that the method achieves a considerable improvement in both recall and precision rate compared with other benchmark methods( random feature subspace,random feature selection based on importance distribution of feature space and linear support vector machine) in sentiment recognition.