Application Research Of Text Sentiment Analysis In Product Reviews

With the rapid development of computer technology,there are a large amount of user comments for particular products on the Internet, such as weibo, blog and e-commerce sites. These comments express the user’s positive or negative attitude directly to product function or performance. Mining these user comments on products is of great significance for enterprises and potential consumers. Based on the background that Xiaomi Tech getting user demand by artificial reading comments is inefficient, this paper mainly applies text sentiment analysis to product reviews mining to enhance the efficiency of work. The research object is Xiaomi mobile phone reviews.First, the paper briefly introduces Xiaomi Tech company and the mode that users participate in research and development, and then analyzes the current situation of the mode and finds it’s ineffective to acquire user demand by manual reading comments, so as to put forward an efficient automation product reviews application solutions. The paper gives the flow chart of product reviews application solutions in detail.Then, we build the product review corpus about Xiaomi phone. After preprocessing data, word segmentation, POS tagging and syntactic analysis, we extract the product features and sentiment words in pairs based on the part of speech template and syntactic relations respectively. Experiments show that these two kinds of methods are efficient. In addition, building a sentiment word dictionary, we have a sentiment classification about product feature based on the dictionary, and show the result in the way of graph.Finally, designing and implementing a product review mining system. The system can capture comments about the specified page, and the main function is extracting product features and emotional words in pairs, sentiment classification and generating a host of visual representations of the data.