Product Attributes Extraction And Visual Representation Of Its Emotion Based On Product Reviews

With the development of Internet commerce, online shopping has become part of people’s daily lives. Much related information hid in the buyers, comment on the goods, which has important reference value for the users, merchants and productors. However, most of the effective information drowned in the vast amounts of the invalid reviews, making the user not get the useful information at a glance. Therefore, how to mining useful information accurately and effectively from these reviews became a research hotspot in recent years.In this thesis, we mainly research the goods attribute words extraction technology and the emotional tendency analysis algorithm towards the goods comments. This main research content of this article is as follows:1. Based on the feature of the comment, we implemented a goods attribute words extraction and sorting algorithm, to obtain the attribute information from the comments.2. Based on the emotional dictionary and machine learning method,we obtained the emotional polarity of the attribute.3. In combination with the product attribute words and their emotional polarity, we designed the visual presentation method in different dimension, from which we could get a intuitive evaluation results of each attribute.