情感分析在电影推荐系统中的应用

发布时间：2018-01-11 01:29

本文关键词：情感分析在电影推荐系统中的应用　出处：《计算机工程与应用》2016年10期 　论文类型：期刊论文

【摘要】：基于用户的协同过滤推荐算法是通过分析用户行为寻找相似用户的集合,其核心是用户兴趣模型的建立以及用户间相似度的计算。传统的用户推荐算法是根据用户评分或者物品信息等行为数据进行个性化推荐,准确率比较低。充分考虑在线评论对于用户之间兴趣相似度的作用,通过对评论的情感分析,构建准确的用户兴趣模型,若用户在评论中表现出来的相似度越高,则表示用户之间的兴趣越相似。实验表明,和传统的基于用户的协同过滤推荐算法相比,基于评论情感分析的协同过滤推荐算法,无论准确率还是召回率都有明显提高。
[Abstract]:The collaborative filtering recommendation algorithm based on user is to find the set of similar users by analyzing the user behavior. The core of this algorithm is the establishment of user interest model and the calculation of user similarity. The traditional user recommendation algorithm is based on the behavior data such as user score or item information. The accuracy is low. Fully consider the role of online reviews to the interest similarity between users, through the emotional analysis of the comments, build an accurate user interest model. The higher the similarity of users in comments, the more similar the interest between users. Experiments show that compared with the traditional collaborative filtering recommendation algorithm based on users. The collaborative filtering recommendation algorithm based on comment emotion analysis can improve both the accuracy and recall rate.
【作者单位】：中国科学技术大学信息科学与技术学院;
【基金】：国家科技支撑计划课题(No.2012BAH73F02) 安徽省科技攻关项目(No.1301b042012)
【分类号】：TP391.3
【正文快照】： 1引言随着互联网的日益普及,各种各样的网站层出不穷,从类似淘宝的购物网站的出现,到美团网等团购网站的大肆崛起,互联网正一步步改变着我们的生活,方便快捷的同时,带给人们的是琳琅满目的商品和服务,随着商品和服务的种类和数量增多,关于它们的网上评论数量也是迅速增加。如

【参考文献】