«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

27

期数:

2017年09期

页码:

64-69

栏目:

智能、算法、系统工程

出版日期:

2017-09-10

文章信息/Info

Title:

 A Recommendation Algorithm of Location Service Based on User Interest

文章编号:

1673-629X（2017)09-0064-06

作者:

 邰淳亮[1] ; 谢怡[1];  孙知信[1][2]

1. 南京邮电大学 计算机学院；2.南京邮电大学 物联网学院

Author(s):

 TAI Chun-liang[1] ; XIE Yi[1] ; SUN Zhi-xin[1][2]

关键词:

 ; 位置服务; 用户建模; 个性化推荐; 用户兴趣

Keywords:

 ; location-based service; user modeling; personalized recommendation; user interest

分类号:

TP301.6

文献标志码:

A

摘要:

 随着移动网络经济用户数、商户数和覆盖范围的扩大,基于用户位置的商家信息推送势必会经历"信息爆炸"和"信息过载",而解决因信息过载导致用户与商家之间的信息迷失的最有效办法是基于用户兴趣的用户位置服务(LBS)推荐.为此,在分别从位置服务技术、用户的兴趣模型,以及个性化推荐算法三个方面深入研究移动位置服务的个性化推荐系统、分析个性化推荐研究现状以及对比分析各种推荐算法的基础上,基于贝叶斯理论,提出了一种适用于移动位置服务的个性化推荐算法.该算法能准确地预测用户在某一情景下消费的兴趣偏好.为验证所提出算法的有效性和可行性,基于所构建的推荐系统进行了实验验证测试.实验结果表明,所提出的算法能够有效地向移动用户提供个性化推荐服务.

Abstract:

 With the increase of consumers and merchants,information pushing services based on consumers’ localities are deemed to go through information explosion and communication overload. Locality service recommendation based on consumers’ interests is an effec-tive method to deal with information misleading between consumers and merchants stemmed from the information overload. In order to solve the problem mentioned above,after the individual recommendation system based on mobile locality has been researched on three as-pects,locality service technique,consumers’ interest model and individual recommendation algorithms respectively as well as the status of investigation on individual recommendation and contrast analysis on various recommendation algorithms. Based on Bayesian theory,an in-dividual recommendation algorithm suitable for mobile location services is proposed,which can predict consumers’ consumption interest accurately in a certain scenario. In order to verify its effectiveness and feasibility,a series of simulation experiments for verification have been conducted with the established recommendation system,which show that it has provided efficient individual recommendation services to mobile consumers.

常用功能

更新日期/Last Update: 2017-10-20