Research On Books Recommendation Algorithm Based On Association Rule And Sentiment Analysis

In today’s digital age, the rapid development of the Internet so that Chinese netizens reached 668 million people, of which about half of the people prefer online comments, but most people like to make their articles, books, web sites, etc. collection, not ugly each user of the information collection is very representative. So how to dig out from these data to help people buy books related information has become one of the hotspots of current research. In this paper, the use of data mining technology to analyze a user’s collection of user habits and emotional expression behavior, giving the user a better book recommendation service.Association rules algorithm is more commonly used in the field of data research methods, often used to discover relationships between different sets of projects, has a very important role in the direction of the recommendation system. Based on Apriori algorithm, an improved method of association rules for analyzing user’s favorite books habits. By analyzing the "Favorites" record "watercress" Web site every reader, the database for each customer collection books analyzing information in order to get all the books frequent item sets and association rules.Association Rules algorithm uses only recommend and can not achieve the best results, this paper based on the context of the book reviews the emotional content analysis to dig out the book review the emotional tendency of readers really think for analysis. Methods for analysis book review section is to first use crawlers obtained watercress users Book Review corpus, and then use the resulting analysis algorithm to analyze the original corpus context-based book review, book review to give emotional bias.The main contents are as follows:(1) association rules algorithms, text mining technology and the status of development at home and abroad were summarized.(2) Based on the classical algorithm Apriori association rules algorithm, an improved algorithm.(3) proposed a context-based text sentiment analysis algorithm.(4) made according to the algorithm, used in books recommended, the experimental results of algorithm analysis.Combined with association rules Bibliographical its consumer reviews sentiment analysis can be recommended to the user more accurate book recommendations. Algorithm provided herein will give more users recommend books by other users, to recommend more effective results.The main innovation of this paper is as follows:(1) Apriori algorithm, an improved algorithm of association rules, reducing the number of calculations, a reduction in time complexity, so as to achieve the purpose to enhance the computing speed, computing books in part from the association rules the important role.(2) in the direction of increasing the recommended books feature comments sentiment analysis, we propose a context-based sentiment analysis algorithm, the correctness of the results of such analysis has improved the accuracy of the entire book recommendation system has also been improved accordingly, improve the user experience.