基于在线评论情感分析的改进协同过滤推荐算法

摘要

针对在线中文评论中用户主观意见的不确定性,提出了一种基于不确定理论的情感分析模型,并结合情感分析模型设计了一种个性化推荐算法.首先,采用分词工具 ICTCLAS和 IKAnalyzer预处理在线中文评论,并基于情感词典(HowNet)计算特征词的点互信息值;然后,应用不确定变量与不确定集设计情感分析模型;最后,根据情感分析模型设计搜索K最近邻居的新方法,并产生推荐.实验结果表明,该方法能够有效提高推荐的准确率,缓解数据稀疏问题.%Aiming at the uncertainty of users'subject opinions in online Chinese review,a sentiment analysis model based on uncertainty theory was proposed.An individual recommendation algorithm was designed on the basis of the proposed sentiment analysis model.First,the tokenizers of ICTCLAS and IKAnalyzer was used to preprocess online Chinese review to generate characteristic words,and the point mutual information value of characteristic words accoun-ting for the sentiment direction were computed based on HowNet dictionary.Then,the sentiment analysis model was established via uncertainty theory of uncertain variables and uncertain sets.In addition,the new similarity formula based on the proposed model was used to search the K-nearest neighbors.Finally,the recommendation lists were given.Experi-mental results show that the proposed method can effectively improve the accuracy of recommendation and alleviate the sparse data problem.