Research On Personalized Recommendation Method Based On Fine-grained Opinion Mining

With the rapid development of the Internet and rapid expansion of e-commerce,online purchase of goods has become a modern lifestyle.Internet shopping platform provides a wide range of products,however,it always takes a lot of time and much effort for users to select products that meet their needs.The emergence of personalized recommendation technology provides a very effective solution to this problem by calculation,and it can imitate the sales staff to help users choose their favorite products.The current personalized recommendation system is generally based on product rating data and user's historical behavioral information,but a large number of valuable information hiding in the product reviews is often overlooked.On the one hand,the rating given by the user is an overall impression of the purchased product which can not meet the needs of other consumers for different evaluation information acquisition of the various aspects of the product.On the other hand,the current research on the use of the commentary text only uses opinion mining and emotion analysis to give consumers a judgment on the integrity of the commodity,but does not combine the ratings and reviews from consumers for the various aspects of the commodity with the personalized recommendation.Based on the above reasons,this paper presents a personalized recommendation method according to fine-grained view mining.In this paper,we obtain the emotion intensity and emotion tendency of various characters of commodities by mining the fine-grained view of the text of user's comments and use the k-means clustering method and DS evidence theory to deal with the results of fine-grained view mining to get the last rating,thereby providing personalized recommendation services for users.Based on the above method to provide recommendation for consumers can be convinced,and is more suitable for consumers' personalized demand.