融合时间特征的社交媒介用户影响力分析

doi:10.6040/j.issn.1671-9352.0.2017.371

山东大学学报（理学版） ›› 2018, Vol. 53 ›› Issue (3): 1-12.doi: 10.6040/j.issn.1671-9352.0.2017.371

• • 下一篇

融合时间特征的社交媒介用户影响力分析

廖祥文^1,2,张凌鹰^1,2,魏晶晶³,桂林^1,2*,程学旗⁴,陈国龙^1,2

1. 福州大学数学与计算机科学学院, 福建福州 350116;2. 福州大学福建省网络计算与智能信息处理重点实验室, 福建福州 350116;3. 福建江夏学院电子信息科学学院, 福建福州 350108;4. 中国科学院网络数据科学与技术重点实验室, 北京 100190

收稿日期:2017-07-31 出版日期:2018-03-20 发布日期:2018-03-13
通讯作者: 桂林(1987— ), 男,讲师,博士,研究方向为情感分析与机器学习. E-mail: guilin.nlp@gmail.com E-mail:liaoxw@fzu.edu.cn
作者简介:廖祥文(1980— ), 男,副教授,博士,研究方向为Web信息检索与观点挖掘. E-mail: liaoxw@fzu.edu.cn
基金资助:
国家自然科学基金资助项目(U1605251,61772135);中国科学院网络数据科学与技术重点实验室开放基金课题资助项目(CASNDST201606);可信分布式计算与服务教育部重点实验室主任基金资助项目(2017KF01);福建省自然科学基金资助项目(2017J01755);赛尔网络下一代互联网技术创新项目(NGII20150901)

User influence analysis of social media with temporal characteristics

LIAO Xiang-wen^1,2, ZHANG Ling-ying^1,2, WEI Jing-jing³, GUI Lin^1,2*, CHENG Xue-qi⁴, CHEN Guo-long^1,2

1. College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350116, Fujian, China;
2. Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, Fujian, China;
3. College of Electronics and Information Science, Fujian Jiangxia University, Fuzhou 350108, Fujian, China;
4. Key Laboratory of Network Data Science and Technology, Chinese Academy of Sciences, Beijing 100190, China

Received:2017-07-31 Online:2018-03-20 Published:2018-03-13

摘要/Abstract

摘要： 针对现有张量影响力模型未能充分考虑用户的时间特征以及在线学习等问题,提出了一种融合时间特征的社交媒介用户影响力分析方法。该方法首先将用户观点、活跃度、网络中心度等特征加入张量模型中,并将张量分解过程中的用户潜在特征矩阵加入时间特征约束;其次,采用随机梯度下降的方法进行张量的分解;最后,通过融合不同张量片的影响力信息得到用户影响力得分。该方法的优点是能够快速分解张量并准确预测特定话题领域下的用户社会影响力,同时能够在已有模型参数的基础上进行新数据的在线训练。实验结果表明,与现有TwitterRank、OOLAM、受限非负张量分解模型等相比, 该方法在平均预测准确率上提升了2%~6%。同时,该方法的时间消耗仅为受限非负张量分解模型的30%~50%。

关键词: 时间特征, 张量, 社会影响力

Abstract: Since both the temporal characteristics and online learning are not fully considered in exsiting tensor influence models, a novel method with temporal characteristics is proposed in this paper. This method constructs tensor with users opinion, activity and network centrality information. Then, a factorizes tensor with stochastic gradient descent algorithm which is constrained by temporal characteristics matrix is deployed in our model. Base on these two steps above, this method calculates user influence by combining different slices of tensor in the end. The advantages of this method are that it can decompose tensor efficiently and satisfy the need of online learning. Experimental results show that the average accuracy of the proposed method is 2% to 6% better than the baseline method such as TwitterRank, OOLAM and constrained nonnegative tensor factorization method. Besides, the running time of the proposed method is only 30% to 50% of constrained nonnegative tensor factorization method.

Key words: Temporal Characteristics, Social Influence, Tensor

中图分类号:

TP391

廖祥文,张凌鹰,魏晶晶,桂林,程学旗,陈国龙. 融合时间特征的社交媒介用户影响力分析[J]. 山东大学学报（理学版）, 2018, 53(3): 1-12.

LIAO Xiang-wen, ZHANG Ling-ying, WEI Jing-jing, GUI Lin, CHENG Xue-qi, CHEN Guo-long. User influence analysis of social media with temporal characteristics[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2018, 53(3): 1-12.

参考文献

[1] 吴信东, 李毅, 李磊. 在线社交网络影响力分析[J]. 计算机学报, 2014, 37(4): 735-752.
[2] WENG J, LIM E P, JIANG J, et al. TwitterRank: finding topic-sensitive influential twitterers[C] //International Conference on Web Search and Web Data Mining. New York: ACM, 2010: 261-270.
[3] BADASHIAN A S, STROULIA E. Measuring user influence in GitHub: the million follower fallacy[C] //Proceedings of the 3rd International Workshop on CrowdSourcing in Software Engineering. New York: ACM, 2016: 15-21.
[4] LEE R K-W, LIM E P. Measuring user influence, susceptibility and cynicalness in sentiment diffusion[C] //Advances in Information Retrieval-37th European Conference on IR Research. Cham: Springer International Publishing, 2015: 411-422.
[5] LI D, TANG J, DING Y, et al. Topic-level opinion influence model(TOIM): An investigation using tencent microblogging[J]. Computer Science, 2015, 6(12): 2657-2673.
[6] CHEN C, GAO D, LI W, et al. Inferring topic-dependent influence roles of Twitter users[C] //Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval. New York: ACM, 2014: 1203-1206.
[7] 魏晶晶, 陈畅, 廖祥文, 等. 基于受限非负张量分解的用户社会影响力分析[J]. 通信学报, 2016, 37(6): 154-162.
[8] EMBAR V R, BHATTACHARYA I, PANDIT V, et al. Online topic-based social influence analysis for the wimbledon championships[C] //Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2015: 1759-1768.
[9] WANG J, LIU Z, ZHAO H. Topic oriented user influence analysis in social networks[C] //IEEE/Wic/ACM International Conference on Web Intelligence and Intelligent Agent Technology. Washington, DC: IEEE Computer Society, 2015: 123-126.
[10] ZAMPARAS V, KANAVOS A, MAKRIS C. Real time analytics for measuring user influence on twitter[C] //IEEE International Conference on TOOLS with Artificial Intelligence. Washington, DC: IEEE Computer Society, 2015: 591-597.
[11] 毛佳昕, 刘奕群, 张敏, 等. 基于用户行为的微博用户社会影响力分析[J]. 计算机学报, 2014, 37(4): 791-800.
[12] HU X, TANG L, TANG J, et al. Exploiting social relations for sentiment analysis in microblogging[C] //Proceedings of the sixth ACM international conference on Web search and data mining. New York: ACM, 2013: 537-546.
[13] KEMPE D, KLEINBERG J, TARDOS É. Maximizing the spread of influence through a social network[J]. Progressive Research, 2008:137-146.
[14] CAI K, BAO S, YANG Z, et al. OOLAM: an opinion oriented link analysis model for influence persona discovery[C] //Proceedings of the fourth ACM international conference on Web search and data mining. New York: ACM, 2011: 645-654.
[15] CUI P, WANG F, YANG S, et al. Item-level social influence prediction with probabilistic hybrid factor matrix factorization[C] //AAAI Conference on Artificial Intelligence. Menlo Park, CA: AAAI Press, 2011.
[16] RODRIGUEZ M G, LESKOVEC J, KRAUSE A. Inferring networks of diffusion and influence[C] //Acm Knowledge Discovery & Data Mining. New York: ACM, 2011: 1019-1028.
[17] KOLDA T G, BADER B W. Tensor decompositions and applications[J]. Siam Review, 2005, 66(4): 294-310.
[18] MAEHARA T, HAYASHI K, KAWARABAYASHI K-I. Expected tensor decomposition with stochastic gradient descent[C] //Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. Menlo Park, CA: AAAI Press, 2016: 1919-1925.
[19] ACAR E, DUNLAVY D M, KOLDA T G. A scalable optimization approach for fitting canonical tensor decompositions[J]. Journal of Chemometrics, 2011, 25(2): 67-86.
[20] ZEILER M D. ADADELTA: an adaptive learning rate method[J]. CoRR, 2012, abs/1212.5701(1): 1-6.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

融合时间特征的社交媒介用户影响力分析

User influence analysis of social media with temporal characteristics

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

多维度评价

本文评价

推荐阅读 0

[1]	于金彪,任永强,曹伟东,鲁统超,程爱杰,戴涛. 可压多孔介质流的扩展混合元解法[J]. 山东大学学报（理学版）, 2017, 52(8): 25-34.
[2]	邢海云,赵建立. 变异机制在网络演化博弈中的应用[J]. 山东大学学报（理学版）, 2016, 51(12): 103-107.
[3]	葛美侠, 李莹, 赵建立, 邢海云. 网络演化博弈的策略一致性[J]. 山东大学学报（理学版）, 2015, 50(11): 113-118.
[4]	邓磊, 赵建立, 刘华, 李莹. k-值控制网络的可控性与可观性[J]. 山东大学学报（理学版）, 2015, 50(04): 27-35.
[5]	赵娜. 实偏对称矩形张量的E-奇异值[J]. J4, 2012, 47(10): 34-37.
[6]	程代展,赵寅，徐相如. 混合值逻辑及其应用[J]. 山东大学学报（理学版）, 2011, 46(10): 32-44.
[7]	孙云1,唐绪兵1,黄时中2. 塞曼哈密顿在\|³PJM_J〉表象中的矩阵元[J]. J4, 2011, 46(1): 56-66.
[8]	毕晓冬. 左拟正规带的张量积[J]. J4, 2009, 44(8): 39-41.
[9]	宋彩芹赵建立王晓东. 三矩阵左半张量积的加权Moore-Penrose[J]. J4, 2009, 44(10): 80-86.
[10]	宋彩芹,赵建立,李东方 . 矩阵左半张量积的(T,S,2)-逆的反序律[J]. J4, 2008, 43(6): 71-76 .
[11]	凌思涛程学汉魏木生. 一般线性四元数矩阵方程的Hermite解[J]. J4, 2008, 43(12): 1-4.