移动个性化应用推荐系统的设计和实现

发布时间：2018-04-21 13:32

本文选题：隐式反馈 + 个性化推荐　；参考：《北京邮电大学》2015年硕士论文

【摘要】：随着移动互联网的飞速发展,移动应用的日益增多,应用市场竞争日益激烈。面对大量的移动应用,移动用户需要花费大量的时间来选择他们感兴趣的应用,开发者也需要花费高昂的推广成本来获取用户。为了便于用户更方便的找到想要的应用,并且能够提供更多的应用曝光机会来帮助移动应用开发者推广应用,降低推广成本,移动个性化应用推荐系统应运而生。本文在传统推荐系统的基础上,结合了移动互联网的特性,设计并实现了一个适用于移动互联网的应用推荐系统。本文首先介绍了课题背景、研究内容以及移动个性化应用推荐系统使用到的相关的技术。其次结合了移动互联网的特点,针对移动终端用户和移动应用开发者进行了需求分析,并提出了一种交叉互推的应用推荐方式。接着本文根据移动终端用户的使用行为和使用习惯,针对传统Slope One推荐算法的不足,提出了一种基于用户隐式反馈的混合推荐算法,并通过测试证明混合推荐算法具有更高的准确性。之后基于当前主流移动终端操作系统的特性,提出了适用于移动个性化应用推荐系统的用户标识方法,并根据对移动用户和移动开发者的需求分析,基于用户隐式反馈的混合推荐算法设计了具有正确性、可用性以及可扩展性的移动个性化应用推荐系统,实现了推荐引擎模块、日志收集处理模块以及离线计算模块。最后对实现的系统进行了功能测试和性能测试,并对本文的工作进行了总结和展望。
[Abstract]:With the rapid development of mobile Internet, the increasing number of mobile applications, application market competition is becoming increasingly fierce. In the face of a large number of mobile applications, mobile users need to spend a lot of time to choose the application they are interested in, and developers also need to spend high cost to get users. In order to facilitate users to find the application they want, and to provide more application exposure opportunities to help mobile application developers promote the application, reduce the promotion costs, mobile personalized application recommendation system came into being. Based on the traditional recommendation system and the characteristics of mobile Internet, this paper designs and implements an application recommendation system for mobile Internet. This paper first introduces the background, research content and the relevant technologies used in mobile personalized application recommendation system. Secondly, combining the characteristics of mobile Internet, this paper analyzes the requirements of mobile terminal users and mobile application developers, and proposes a cross-push application recommendation method. Then, according to the usage behavior and usage habits of mobile terminal users and the shortcomings of traditional Slope One recommendation algorithm, a hybrid recommendation algorithm based on user implicit feedback is proposed in this paper. The test results show that the hybrid recommendation algorithm has higher accuracy. Then, based on the characteristics of the current mainstream mobile terminal operating system, a user identification method suitable for mobile personalized application recommendation system is proposed, and according to the needs of mobile users and mobile developers, Based on the user implicit feedback hybrid recommendation algorithm, a mobile personalized application recommendation system with correctness, availability and extensibility is designed. The recommendation engine module, log collection and processing module and off-line computing module are implemented. Finally, the function test and performance test of the system are carried out, and the work of this paper is summarized and prospected.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.3

【参考文献】