移动商务中融合签到位置与用户间相似性的兴趣点精准推荐

向位置频繁变化的移动用户进行精准推荐,如何提高推荐准确性,已经成为一个理论研究与实践中的热点与难点问题.针对于此,本文提出了面向移动社交商务的精准用户兴趣点推荐模型--MRGR.首先,利用用户的历史签到信息,通过改进核密度估计对兴趣点进行预测;其次,进一步考虑到用户间的签到相似性,使用信息熵定量表示用户移动的随机性和不确定性;最后,融合地理位置信息与用户间相似性,精准推荐用户兴趣点,并在Foursquare数据集上进行验证.实验结果表明:与传统模型相比,提出的模型在准确预测用户兴趣点的同时,可以有效缓解数据稀疏性和冷启动问题,并在准确率和召回率上都取得了显著的提高.成果将为移动商务中,如何更好满足企业的精准推送与用户个性化需求提供有力的技术支持和决策服务.

Recommendation of POI by integrating user similarity and location information in mobile commerce

It has become a hot and difficult problem in the theoretical research to improve the accuracy of recommendation of mobile users with frequent changes of position. This paper proposes a mobile user interest group of point-of-interests (POI) recommendation model-MRGR for mobile social commerce. The kernel density estimation is improved to predict the interest points; secondly, the similarity between users' check-in is further considered, and the information entropy is used to quantify the randomness and uncertainty of the user's movement. Finally, the similarity between the geographic location information and the user is fused and verified on the Foursquare dataset. The experimental results show that the model proposed in this paper can be compared with the traditional model. It can effectively alleviate the problem of data sparsity and cold start, accurately predict the user's interest points, and have made significant improvements in accuracy and recall. The results will provide strong technical support and decision-making service for the mobile commerce on how to better meet the precise service of the enterprise and the personalized needs of the users.