基于LDA模型的餐厅推荐方法研究

Research on Recommendation Method of Restaurant Based on LDA Model

随着网络的飞速发展,餐饮类的评价信息数量急剧增加。对餐饮评价进行有效分析不仅能够帮助消费者进行用餐选择,还可以帮助商家对餐厅服务进行改进。为此,提出了一种基于LDA(Latent Dirichlet Allocation)模型的餐厅推荐方法。首先,对餐厅评价信息进行情感分类,获取积极评价和好评率;其次,根据LDA模型对积极评价信息文本进行聚类,生成餐厅标签;最后,计算用户需求与餐厅标签的相似度,根据相似度和好评率向用户推荐餐厅。基于通过网络获取的真实餐饮评价信息进行实验,结果表明,该方法生成的餐厅标签的效果好,能准确地向用户推荐餐厅。

With the rapid development of the network,the amount of the evaluation information of the food and beverage has increased dramatically.The effective analysis of the evaluation information can not only help the consumers choose the suitable restaurant,but also help the businesses improve service.For this purpose,a restaurant recommendation method based on LDA（Dirichlet Allocation Latent）model was proposed.First of all,it classifies the evaluation information according to the emotional tendencies,and then gets the positive evaluation and praise rate.Secondly,it manipulates the LDA model for text clustering to generate restaurant tags.Finally,it calculates the similarity between the user’s needs and the restaurant tags,and according to the similarity and the rate of praise,recommends the suitable restaurants to customers.We got the real food and beverage comments from the Internet,and carried out the experiment.As a result,the effect of the restaurant tags produced from this method is good,which could accurately recommend the restaurants to users.