电信科学 ›› 2014, Vol. 30 ›› Issue (8): 131-135.doi: 10.3969/j.issn.1000-0801.2014.08.019

• 运营技术广角 • 上一篇    下一篇

灵活适应不同业务的个性化推荐系统研究

陶彩霞1,袁海2,陈康1,马安华2   

  1. 1 中国电信股份有限公司广东研究院 广州510630
    2 中国电信股份有限公司江苏分公司 南京210037
  • 出版日期:2014-08-15 发布日期:2014-08-15

Research on Personalized Recommender System Adapting to Different Business

Caixia Tao1,Hai Yuan2,Kang Chen1,Anhua Ma2   

  1. 1 Guangdong Research Institute of China Telecom Co.,Ltd.,Guangzhou 510630, China
    2 Jiangsu Branch of China Telecom Co., Ltd., Nanjing 210037, China
  • Online:2014-08-15 Published:2014-08-15

摘要:

互联网技术的迅速发展,尤其是以个性化为主要特点的Web 2.0的不断成熟,使得大量信息同时呈现在人们面前,个性化推荐技术的价值日益凸显。提出了一种灵活适应不同业务的个性化推荐系统设计方案,采集并分析用户的所有显式行为和隐式行为,结合用户行为配置文件和熵值法分析用户对物品的兴趣度,并引入时间遗忘函数,解决用户兴趣漂移问题,然后基于协同过滤技术得到用户的个性化推荐列表。通过实际数据进行测试,给出了系统推荐效果评估分析。

关键词: 个性化推荐系统, 用户兴趣分析, 熵, 兴趣漂移, 协同过滤

Abstract:

The rapid development of internet technology, especially the developed of Web 2.0 which has the main feature of personality, making a lot of information in front of people at the same time. The value of personalized recommendation technology has become increasingly prominent. A design of personalized recommender system adapting to different business was presented, including the collection and analysis of user's all explicit and implicit behavior, using user behavior profiles and entropy method to determine the weight of the behavior for analyzing user interest, the introduction of time forgetting function to solve the problem of user interest drift, and getting a list of the user's personalized recommendation based on collaborative filtering technology. Finally, the system test assessment analysis based on the actual data was provided.

Key words: personalized recommender system, user interest analysis, entropy, interest drift, collaborative filtering

No Suggested Reading articles found!