作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (10): 60-66,73. doi: 10.19678/j.issn.1000-3428.0055979

• 人工智能与模式识别 • 上一篇    下一篇

面向微博用户的个性化推荐算法研究

周炜翔1, 张雯1, 杨博2, 柳毅2, 张琳2, 张仰森1   

  1. 1. 北京信息科技大学 智能信息处理研究所, 北京 100101;
    2. 国家计算机网络应急技术处理协调中心, 北京 100029
  • 收稿日期:2019-09-11 修回日期:2019-10-29 发布日期:2019-11-08
  • 作者简介:周炜翔(1993-),男,硕士研究生,主研方向为自然语言处理;张雯,硕士研究生;杨博、柳毅、张琳,博士;张仰森(通信作者),教授、博士。
  • 基金资助:
    国家自然科学基金"网络社交媒体中特定社会安全事件的侦测分析与态势评估研究"(61772081)。

Research on Personalized Recommendation Algorithm for Microblog Users

ZHOU Weixiang1, ZHANG Wen1, YANG Bo2, LIU Yi2, ZHANG Lin2, ZHANG Yangsen1   

  1. 1. Institute of Intelligent Information Processing, Beijing Information Science and Technology University, Beijing 100101, China;
    2. National Computer Network Emergency Response Technical Team/Coordination Center of China, Beijing 100029, China
  • Received:2019-09-11 Revised:2019-10-29 Published:2019-11-08

摘要: 微博的个性化推荐对于提升用户体验和帮助用户及时、准确地获取信息具有重要意义。在分析微博用户行为模式的基础上,提出一种基于情景建模和卷积神经网络的微博个性化推荐模型。从时间和地域两个维度对用户进行情景建模,提取用户的时间情景模式和地域情景模式,同时给出情景模式相似度计算方法,对用户的情景模式进行扩展,捕捉用户感兴趣的情景模式倾向,在此基础上建立用户个性化情景模式库,采用卷积神经网络构建个性化微博推荐模型,实现微博用户的个性化推荐。实验结果表明,与ILCAUSR、RA-CD算法相比,该模型具有较好的推荐效果,相比于时间情景模型和地域情景模型,其平均绝对误差和平均用户满意度指标均达到最优效果。

关键词: 个性化推荐, 情景建模, 卷积神经网络, 情景模式库, 用户满意度

Abstract: Personalized recommendation of microblog is crucial to improving user experience and helping users obtain information accurately in time.Based on the analysis of behavior patterns of microblog users,this paper proposes a personalized recommendation model for the microblog based on scenario modeling and Convolutional Neural Network(CNN).Scenario modeling is implemented for users from the dimensions of time and region,so as to extract the user’s temporal scenario pattern and geographical scenario pattern.Then a calculation method of scenario pattern similarity is provided to extend the scenario patterns of users,capturing the scenario pattern tendency that users are interested in.On this basis,a personalized scenario mode library of the user is established,and the CNN is used to construct a personalized recommendation model for microblog users.Experimental results on real data of the microblog show that compared with the ILCAUSR and RA-CD algorithms,the proposed model has better recommendation performance,and achieves the optimal effect in Mean Absolute Error(MAE) and Average User Satisfaction(AUS) indexes compared with the temporal scenario model and geographical scenario model.

Key words: personalized recommendation, scenario modeling, Convolutional Neural Network(CNN), scenario pattern library, user satisfaction

中图分类号: