基于移动用户浏览行为的推荐模型

发布时间：2018-07-26 19:30

【摘要】：推荐算法已经被广泛地应用于很多领域。但是如果利用传统的推荐算法预测移动用户浏览互联网的行为,并在此基础上对移动用户进行个性化的内容推荐,传统推荐算法的推荐效果往往比较差。该文通过分析移动用户浏览互联网的记录,得出传统推荐算法效果差的原因。在此基础上,提出了一个基于移动用户浏览行为的推荐模型,即RMBDMU。该模型可以对移动用户浏览互联网的行为进行预测,在预测的基础上对移动用户进行内容推荐。为了验证推荐模型的有效性,在真实的移动用户浏览互联网的行为数据上进行了实验。实验结果显示基于移动用户浏览行为的推荐模型比传统的推荐算法更为有效。
[Abstract]:Recommendation algorithms have been widely used in many fields. However, if the traditional recommendation algorithm is used to predict the behavior of mobile users browsing the Internet, and on the basis of which mobile users can be personalized content recommendation, the traditional recommendation algorithm often has a poor recommendation effect. By analyzing the records of mobile users browsing the Internet, the reason for the poor performance of the traditional recommendation algorithm is obtained. On this basis, a recommendation model based on the browsing behavior of mobile users, namely RMBDMU, is proposed. The model can predict the behavior of mobile users browsing the Internet and recommend the content of mobile users on the basis of prediction. In order to verify the validity of the recommendation model, experiments were carried out on the real mobile users browsing the Internet behavior data. Experimental results show that the recommendation model based on mobile user browsing behavior is more effective than the traditional recommendation algorithm.
【作者单位】：电子科技大学信息与软件工程学院;电子科技大学网络与数据安全四川省重点实验室;
【基金】：国家自然科学基金(61133016,61300191,61202445,61370026) 四川省科技支撑计划(2014GZ0106,2016JZ0020)
【分类号】：TP391.3

【相似文献】