融合用户自身因素与互动行为的微博用户影响力计算方法

doi:10.11896/jsjkx.181202253

摘要/Abstract

摘要： 由于微博高影响力用户在商品营销、社会舆论引导等方面起着重要的作用,因此挖掘高影响力用户成为了微博社交网络中的热点研究问题。针对微博用户影响力计算中存在交互行为与用户自身因素分析不全面的问题,提出了微博用户影响力计算方法MBUI-SFIM(Micro-blog userinfluence based on user’s self-factors and interaction computing model)。该方法考虑了微博用户直接影响力和间接影响力两个方面:在用户直接影响力计算中,通过对用户的自身因素如微博用户粉丝数、用户活跃度、近期微博质量等的分析,计算出用户的初始影响力,然后分析用户互动行为如用户的微博可见率、微博用户互动系数,计算出用户传播能力,最后将初始影响力与用户传播能力相结合,基于改进PageRank算法计算出用户直接影响力;在用户间接影响力计算中,通过对用户网络图连接结构进行分析,根据不相邻用户连接路径的不同,将用户间接影响具体分为简单路径、重复路径、复杂路径3种情况进行讨论,从而计算出用户间接影响力。实验结果表明,相比PageRank算法和MR-UIRank算法,所提算法在用户排名准确性上分别提高了14.8%和8.3%。

关键词: PageRank, 间接影响力, 交互行为, 微博, 直接影响力, 自身因素

Abstract: Weibo users with high-impact play an important role in commodity marketing and social publicity guidance,so mining high-impact users becomes a hot research issue in Weibo social networks.As for the problems of incomplete behavior analysis of interaction behavior and user’s own factors in calculation of micro-blog user influence,the micro-blog user influence based on user’s self-factors and interactiont computing model was proposed.This method considers the direct influence and indirect influe-nce of Weibo users.In the user’s direct influence calculation phase,the initial influence of the user is calculated by analyzing the user’s own factors such as the number of fans of Weibo users,user activity,and recent microblog quality.Then the user interaction behavior is analyzed,such as the user’s microblog visibility rate,microblog user interaction coefficient,so as to calculate the user communication ability.Finally,by combining the initial influence with the user communication ability,the user’s direct influe-nce is cakulated based on the improved PageRank algorithm.In the calculation of user indirect influence phase,through the analysis of the connection structure of the user network diagram and according to the different connection paths of non-adjacent users,the indirect impact of the user is divided into three categories:simple path,repeated path and complex path,then the user indirect influence is calculated.The experimental results show that the proposed algorithm is 14.8% and 8.3% higher than the PageRank algorithm and the MR-UIRank algorithm in terms of the user ranking accuracy.

Key words: Direct influence, Indirect influence, Interaction behavior, Microblog, PageRank, Self-factor

中图分类号:

TP393

王新胜,马树章. 融合用户自身因素与互动行为的微博用户影响力计算方法[J]. 计算机科学, 2020, 47(1): 96-101. https://doi.org/10.11896/jsjkx.181202253

WANG Xin-sheng,MA Shu-zhang. Method of Weibo User Influence Calculation Integrating Users’ Own Factors and Interaction Behavior[J]. Computer Science, 2020, 47(1): 96-101. https://doi.org/10.11896/jsjkx.181202253

参考文献

[1]BINGOL K,ERAVCI B,ETEMOGLU C O,et al.Topic-based influence computation in social networks under resource constraints[J].IEEE Transactions on Services Computing,2016,12(6):970-986.
[2]LIU Q,XIANG B,YUAN N J,et al.An influence propagation view of pagerank[J].ACM Transactions on Knowledge Discovery from Data (TKDD),2017,11(3):1-30.
[3]PENG S,YANG A,CAO L,et al.Social influence modeling using information theory in mobile social networks[J].Information Sciences,2017,379:146-159.
[4]WENG J,LIM E P,JIANG J,et al.Twitterrank:finding topic-sensitive influential twitterers[C]∥Proceedings of the Third ACM International Conference on Web Search and Data Mining.ACM,2010:261-270.
[5]SUN H,ZUO T.Research and Realization of Influence Optimization in Cloud Computing Environment[J].Journal of Chinese Computer Systems,2018,39(1):42-47.
[6]PAGE L,BRIN S,MOTWANI R,et al.The PageRank citation ranking:Bringing order to the Web[J].Stanford Digital Lib-raries Working Paper,1998,9(1):1-14.
[7]YAMAGUCHI Y,TAKAHASHI T,AMAGASA T,et al. Turank:Twitter user ranking based on user-tweet graph analysis[C]∥International Conference on Web Information Systems Engineering.Springer,Berlin,Heidelberg,2010:240-253.
[8]MEEYOUNG C,HAMED H,FABRICIO B,et al.Measuring user influence in twitter:The million follower fallacy[C]∥Fourth International AAAI Conference on Weblogs and Social Media.Menlo Park:AAAIPress,2010:10-17.
[9]LI G L,CHU Y P,FENG J H,et al.Influence maximization on multiple social networks [J].Chinese Journal of Computers,2016,39:643-656.
[10]LIU L,TANG J,HAN J,et al.Mining topic-level influence in heterogeneous networks[C]∥Proceedings of the 19th ACM International Conference on Information and Knowledge Management.ACM,2010:199-208.
[11]MAO J X,LIU Y Q,ZHANG M,et al.Social influence analysis for micro-blog user based on user behavior[J].Chinese Journal of Computers,2014,37(4):791-798.
[12]ZHOU J,ZHANG Y,WANG B,et al.Predicting user influence in microblogs[C]∥2016 First IEEE International Conference on Computer Communication and the Internet (ICCCI).IEEE,2016:292-295.
[13]WU L,YANG B,JIAN M,et al.MPPR:Multi Perspective Page Rank for User Influence Estimation[C]∥2018 IEEE International Conference on Big Data and Smart Computing (BigComp).IEEE,2018:25-29.
[14]BAKSHY E,HOFMAN J M,MASON W A,et al.Everyone’s an influencer:quantifying influence on twitter[C]∥Proceedings of the Fourth ACM International Conference on Web Search and Data Mining.ACM,2011:65-74.
[15]LI C,XIONG F.Social recommendation with multiple influence from direct user interactions[J].IEEE Access,2017,5:16288-16296.
[16]YANG L U,HUAKANG L I,GUOZI S.Distributed microblog crawler system based on P2P[J].Journal of Jiangsu University,2016,37(3):296-301.
[17]WEN Z,KVETON B,VALKO M,et al.Online influence maximization under independent cascade model with semi-bandit feedback[C]∥Advances in Neural Information Processing Systems.2017:3022-3032.
[18]KEMPE D,KLEINBERG J,TARDOS É.Maximizing the spread of influence through a social network[C]∥Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.ACM,2003:137-146.

相关文章 15

[1]	谢柏林, 黎琦, 邝建. 基于隐半马尔可夫模型的微博流行信息检测方法 Microblog Popular Information Detection Based on Hidden Semi-Markov Model 计算机科学, 2022, 49(6A): 291-296. https://doi.org/10.11896/jsjkx.210800011
[2]	徐建民, 孙朋, 吴树芳. 传播路径树核学习的微博谣言检测方法 Microblog Rumor Detection Method Based on Propagation Path Tree Kernel Learning 计算机科学, 2022, 49(6): 342-349. https://doi.org/10.11896/jsjkx.210400096
[3]	李玉强, 张伟江, 黄瑜, 李琳, 刘爱华. 基于高斯分布的改进词嵌入主题情感模型 Improved Topic Sentiment Model with Word Embedding Based on Gaussian Distribution 计算机科学, 2022, 49(2): 256-264. https://doi.org/10.11896/jsjkx.201200082
[4]	史伟, 付月. 考虑语境的微博短文本挖掘:情感分析的方法 Microblog Short Text Mining Considering Context:A Method of Sentiment Analysis 计算机科学, 2021, 48(6A): 158-164. https://doi.org/10.11896/jsjkx.210200089
[5]	郁友琴, 李弼程. 基于多粒度文本特征表示的微博用户兴趣识别 Microblog User Interest Recognition Based on Multi-granularity Text Feature Representation 计算机科学, 2021, 48(12): 219-225. https://doi.org/10.11896/jsjkx.201100128
[6]	王晓涵, 谭陈琛, 相艳, 余正涛. 基于双嵌入卷积神经网络的涉案微博评价对象抽取 Aspect Extraction of Case Microblog Based on Double Embedded Convolutional Neural Network 计算机科学, 2021, 48(12): 319-323. https://doi.org/10.11896/jsjkx.201100105
[7]	潘培贤, 邹兆年, 李发明. 历史图上基于CSR结构的PageRank算法 CSR-based PageRank Algorithm on Historical Graphs 计算机科学, 2020, 47(9): 88-93. https://doi.org/10.11896/jsjkx.190800122
[8]	袁得嵛, 章逸钒, 高见, 孙海春. 基于用户特征提取的新浪微博异常用户检测方法 Abnormal User Detection Method in Sina Weibo Based on User Feature Extraction 计算机科学, 2020, 47(6A): 364-368. https://doi.org/10.11896/JsJkx.190700008
[9]	刘宇东, 孙豪, 蒋运承. 融合内容相似度与多特征计算的个性化微博推荐模型 Personalized Microblog Recommendation Model Integrating Content Similarity and Multi-feature Computing 计算机科学, 2020, 47(10): 97-101. https://doi.org/10.11896/jsjkx.190700073
[10]	刘小捷, 吕晓强, 王晓玲, 张伟, 赵安. 基于维基百科类别图的推特用户兴趣挖掘 Mining User Interests on Twitter Using Wikipedia Category Graph 计算机科学, 2019, 46(9): 79-84. https://doi.org/10.11896/j.issn.1002-137X.2019.09.010
[11]	周艳芳, 周刚, 鹿忠磊. 一种基于迁移学习及多表征的微博立场分析方法 Approach of Stance Detection in Micro-blog Based on Transfer Learning and Multi-representation 计算机科学, 2018, 45(9): 243-247. https://doi.org/10.11896／j.issn.1002-137X.2018.09.040
[12]	刘慧婷, 程雷, 郭孝雪, 赵鹏. 实时个性化微博推荐系统 Real-time Personalized Micro-blog Recommendation System 计算机科学, 2018, 45(9): 253-259. https://doi.org/10.11896／j.issn.1002-137X.2018.09.042
[13]	陈志雄,王时绘,高榕. 基于情感倾向性分析的微博意见领袖识别模型 Recognition Model of Microblog Opinion Leaders Based on Sentiment Orientation Analysis 计算机科学, 2018, 45(5): 168-175. https://doi.org/10.11896/j.issn.1002-137X.2018.05.028
[14]	王嵘冰,安维凯,冯勇,徐红艳. 基于标签和PageRank的重要微博用户推荐算法 Important Micro-blog User Recommendation Algorithm Based on Label and PageRank 计算机科学, 2018, 45(2): 276-279. https://doi.org/10.11896/j.issn.1002-137X.2018.02.047
[15]	王礼敏,严倩,李寿山,周国栋. 基于双通道LSTM模型的用户性别分类方法研究 User Gender Classification with Dual-channel LSTM 计算机科学, 2018, 45(2): 121-124. https://doi.org/10.11896/j.issn.1002-137X.2018.02.021

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed