【摘要】 针对网络商品评论数据不能有效引导买方做出合理选择的问题，提出一种融合LDA主题模型和支持向量机的商品个性化推荐方法。首先爬取不同类型商品的用户评论数据并对其进行预处理；其次建立基于LDA的主题模型并对其特点进行量化；最后利用支持向量机实现商品个性化推荐。以智能手机商品为例进行实验分析，结果表明，所提方法能获得98%以上的分类精度。更多还原

【Abstract】 Aiming at the problem that online commodity review data could not effectively guide buyers to make reasonable choices, a commodity personalized recommendation method integrating LDA topic model and support vector machine was proposed. Firstly, the user comment data of different types of goods were crawled and preprocessed.Secondly, the topic model based on LDA was established and its characteristics was quantified.Finally, support vector machine was used to realize commodity personalized recommendation. The experimental results on smart phone products showed that the proposed method could achieve more than 98% classification accuracy.更多还原