基于个性化情景的移动商务信任推荐模型研究

发布时间：2018-08-17 09:22

【摘要】：【目的/意义】移动互联网时代,移动电子商务用户的个性化信息需求具有极强的情景依赖性与感知信任性,针对目前移动商务信息服务个性化和准确性较低,提出融入用户个性化情景与用户间信任关系的推荐模型。【方法/过程】首先,通过用户当前情景和历史评分数据计算出对每个用户即时信息需求影响最大的K个情景要素,以此构造用户个性化情景,然后结合不同信任环境下的用户信任度矩阵改进已有的不同信任信息环境下用户情景兴趣推荐方法,进而进行项目推荐。【结果/结论】通过Movie lens与Book-Crossing数据集对本文提出的算法和其它两种算法进行比较,实验结果表明:本模型具有较高的推荐准确率,可有效地解决移动商务环境下的个性化推荐问题。
[Abstract]:[Objective/Significance] In the era of mobile Internet, the personalized information needs of mobile e-commerce users have strong situational dependence and perceptual trust. Aiming at the low individuation and accuracy of mobile commerce information services, a recommendation model incorporating user personalized scenarios and trust relationship between users is proposed. Firstly, K scenario elements which have the greatest impact on each user's instant information demand are calculated by user's current scenario and historical score data, so as to construct user's personalized scenario. Then, the user's scenario interest recommendation method under different trust information environments is improved by combining the user's trust matrix under different trust environments, and then the user's scenario interest recommendation method under different trust information environments [Results/Conclusion] Comparing Movie lens and Book-Crossing datasets, the proposed algorithm and the other two algorithms show that the model has high recommendation accuracy and can effectively solve the personalized recommendation problem in mobile commerce environment.
【作者单位】：哈尔滨理工大学管理学院;哈尔滨理工大学高新技术产业发展研究中心;
【基金】：国家自然科学基金(71272191);国家自然科学基金(71672050) 黑龙江省自然科学基金项目(G201301)
【分类号】：F713.36;G252

【相似文献】