基于情境聚类扩展用户画像的旅游景点推荐方法

Research on Scenic Spots Recommendation Method Based on Context Clustering

随着我国旅游业发展和人们对旅游服务需求的不断增多,旅游景点的精准推荐不仅可以节约用户大量的时间,还能够提升用户的消费体验,为此本文提出一种基于情境聚类扩展用户画像的旅游景点推荐方法。基于用户情境数据进行情境聚类,构建出扩展的用户画像模型;根据群体用户画像反映的用户特征,结合情境标签设置个体用户情境特征属性,再融合关联规则和相似度计算为用户推荐个性化的景点。实验结果表明,相较于传统的关联规则Apriori推荐算法或加权Apriori推荐算法,基于情境聚类扩展用户画像的旅游景点推荐方法具有更优的推荐结果。

With the development of China's tourism industry and the increasing demand for tourism services,the accurate recommendation of tourist attractions can not only save users time,but also improve their consumption experience.This paper proposes a recommendation method for tourist attractions based on context clustering to expand user portrait.Based on the us-er situation data,the extended user portrait model is constructed.According to the user characteristics reflected by group user portraits,user context feature attributes are set in combination with context labels,and personalized scenic spots are recom-mended for users by integrating association rules and similarity calculation.The experimental results show that compared with the traditional association rule Apriori algorithm or the weighted Apriori algorithm,the tourist attractions recommendation method based on the context clustering extended user portrait has better recommendation results.