An Improved SVM for Book Review Sentiment Polarity Analysis

摘要

In the internet age, whether a book has the value of reading, online comments play an important role. The data set in this paper is 4,000 comments obtained by the web crawler in Douban Reading. Based on the improved support vector machine (SVM) algorithm, a sentiment analysis has been given to these comments. The experimental results show that the improved SVM algorithm has a good effect on the rate and accuracy of sentiment polarity analysis of book reviews.