摘要
传统推荐模型的性能依赖于用户兴趣特征模型的质量,而用户的兴趣特征难以有效的收集和建模.网络购物评论包含了消费者对商品的各种特征的描述.提出的基于购物评论的商品推荐模型,通过对购物评论的预处理,借助文本分类获得产品特征,依靠本体技术实现对商品特征结构化描述,通过情感分析判断评论内容的极性,最后通过推荐计算模型,生成商品推荐意见.
Traditional recommender model performance depends on characteristics of user interest model quality, but the customer＇s characteristics are difficult to effectively collect and model. Reviews include the consumer online shopping, the various features of the description of goods. The proposed shopping model recommendation of goods, based on reviews, uses the comments of the pre-shopping and text classification for product features, relies on a structured description of goods features by ontology technology, determine the polarity of reviews by using emotional analysis, and finally, generates merchandise recommendations by using the recommendation computational model.
出处
《西南民族大学学报（自然科学版）》
CAS
2012年第3期471-474,共4页
Journal of Southwest Minzu University(Natural Science Edition)
基金
四川省科技厅项目(2011JY0094)
关键词
购物评论
推荐模型
本体
评论挖掘
customer review
recommender model
ontology
review mining
