SPECIAL TOPIC: Evolution and Governance of Social Media Group Behavior

Research on Social Media Community Detection Integrating Semantic Sentiment and Keywords Matching

  • Li Shiyu ,
  • Zhang Xiangxian ,
  • Zhang Yuxing ,
  • Zhu Linlin ,
  • Li He
Expand
  • 1 Pengcheng National Laboratory, Shenzhen 518000;
    2 School of Business and Management, Jilin University, Changchun 130012;
    3 Aerospace Archives, Beijing 100830
Li Shiyu, PhD, Pengcheng Postdoc; Zhang Xiangxian, Professor, PhD, Doctoral Supervisor; Zhang Yuxing, Assistant Librarian, Master’s Degree; Zhu Linlin, PhD, Associate Professor, Master’s Supervisor; Li He, Professor, PhD, Doctoral Supervisor, Corresponding Author, E-mail: lihe200303@163.com.

Received date: 2024-12-03

  Revised date: 2025-04-19

  Online published: 2025-11-20

Supported by

This work is supported by the youth project of National Social Science Fund of China titled, “Research on the Formation Mechanism and Evaluation of the Influence of Online Reviews from the Perspective of Cognitive Schemas” (Grant No. 21CTQ015).

Abstract

[Purpose/Significance] This study proposes a social media community detection method that integrates semantic sentiment analysis and keyword matching. By increasing the number and density of matrix features, the accuracy of user community detection based on matrix decomposition is improved, enabling deeper insights into social media users’ interests and preferences. [Method/Process] The study constructed a user attitude matrix through sentiment analysis and with keyword matching. To enrich the matrix features, users’ common followings on social media as indirect interactions were introduced into the user interaction matrix. The GSNMF (Graph Regularized Semi-Nonnegative Matrix Factorization) algorithm was used to achieve precise user community detection. [Result/Conclusion] The results show that the method combining semantic sentiment and keyword matching techniques significantly improves the accuracy and effectiveness of user community detection. Additionally, by leveraging the community feature matrix obtained through matrix decomposition, the major interests and preferences of different community users can be effectively analyzed.

Cite this article

Li Shiyu , Zhang Xiangxian , Zhang Yuxing , Zhu Linlin , Li He . Research on Social Media Community Detection Integrating Semantic Sentiment and Keywords Matching[J]. Library and Information Service, 2025 , 69(23) : 33 -45 . DOI: 10.13266/j.issn.0252-3116.2025.23.003

References

[1] 张文宏. 网络社群的组织特征及其社会影响[J]. 江苏行政学院学报, 2011(4): 68-73. (ZHANG W H. Organizational characteristics and social impact of network social groups[J]. The Journal of Jiangsu Administration Institute, 2011(4): 68-73.)
[2] HU J, SHAO H, TANG J, et al. Research on reading subject recommendation of library based on data analysis[J]. Journal of physics: conference series, 2020, 1607(1): 012022(7pp).
[3] 卓玛曲松. 探索与发现: 初中数学兴趣小组活动的效果分析[C]//中国智慧工程研究会. 2024数字化教育教学交流会论文集. 长沙: 2024数字化教育教学交流会, 2025: 1-2. (ZHUOMA Q S. Exploration and discovery: Effectiveness analysis of mathematics interest group activities in junior high schools[C]//China Wisdom Engineering Association Proceedings of the 2024 Digital Education and Teaching Exchange Conference. Changsha: the 2024 Digital Education and Teaching Exchange Conference, 2025: 1-2.)
[4] 滕广青, 贺德方, 彭洁, 等. 基于“ 用户— 标签” 关系的社群知识自组织研究[J]. 图书情报工作, 2014, 58(20): 106-111. (TENG G Q, HE D F, PENG J, et al. Study on self-organization of community knowledge based on “user-tag” relationship[J]. Library and information service, 2014, 58(20): 106-111.)
[5] KETS W S A. A belief-based theory of homophily[J]. Games and economic behavior, 2019, 115: 410-435.
[6] 李贺, 刘嘉宇, 沈旺, 等. 社交参与视角下超图增强的学习趣缘社群群体检测研究[J]. 情报学报, 2024, 43(12): 1425-1439. (LI H, LIU J Y, SHEN W, et al. Group detection in interestbased learning communities enhanced by hypergraph from a social engagement perspective[J]. Journal of the China Society for Scientific and Technical Information, 2024, 43(12): 1425- 1439.)
[7] OOMS W, BELL J, KOK R A W. Use of social media in inbound open innovation: building capabilities for absorptive capacity[J]. Creativity and Innovation Management, 2015, 24(1): 136-150.
[8] SAH P, MANN J, BANSAL S. Disease implications of animal social network structure: a synthesis across social systems[J]. Journal of animal ecology, 2018, 87(3): 546-558.
[9] IBRAHEEM S F, AL-SARRAY B. A review of community detection based on modularity optimization in complex networks[J]. Iraqi journal of science, 2024, 65(5): 2775-2793.
[10] 孙赛赛, 何雅迪, 王兰. 青年群体网络“ 圈层化” 的样态透视与应对之策[J]. 理论观察, 2024(4): 95-98. (SUN S S, HE Y D, WANG L. The perspective and countermeasures of the “circle like” network of youth groups [J]. Theoretic observation, 2024(4): 95-98.)
[11] 江文华, 徐健, 李纲, 等. 基于研究兴趣相似性网络的我国图书馆学研究社群分析[J]. 现代情报, 2019, 39(9): 21-27. (JIANG W H, XU J, LI G, et al. Analysis on domestic research community of library science based on research interest similarity network[J]. Journal of modern information, 2019, 39(9): 21-27.)
[12] 邓胜利, 胡吉明. Web 2.0环境下网络社群理论研究综述[J]. 中国图书馆学报, 2010(5): 90-95. (DENG S L, HU J M. Review on online community theory in web2.0 environment[J]. Journal of library science in China, 2010(5): 90-95.)
[13] 董伟, 陶金虎. 基于主题偏好的在线健康社区用户兴趣群体识别研究——以医享网为例[J]. 情报科学, 2021, 39(3): 88- 93, 119. (DONG W, TAO J H. Research on user interest group identification in online health communities based on theme preferences: a case study of Yixiang network[J]. Information science, 2021, 39(3): 88-93, 119.)
[14] 李纲, 唐晶, 毛进, 等. 基于演化事件探测的学科领域科研社群演化特征研究——以图书馆学情报学为例[J]. 图书情报工作, 2021, 65(17): 79-90. (LI G, TANG J, MAO J, et al. Research on the evolution characteristics of scientific research communities in subject fields based on evolutionary event detection: an example of LIS[J]. Library and information service, 2021, 65(17): 79-90.)
[15] WANG J, LIU Z, ZHAO H. Group recommendation using topic identification in social networks[C]//2014 sixth international conference on intelligent human-machine systems and cybernetics. Hangzhou: IEEE, 2014, 1: 355-358.
[16] 张红宇, 王坚强, 马华. 结合社会网络分析和多维特征聚类的学习小组划分方法[J]. 计算机应用研究, 2013, 30(3): 732-735, 741. (ZHANG H Y, WANG J Q, MA H. Grouping approach of learning team based on social network analysis and multidimensional feature clustering [J]. Application research of computers, 2013, 30(3): 732-735, 741.)
[17] 何军, 刘业政, 王锦坤. 基于社会选择和社会影响的社交网络社群分类与群推荐策略研究[J]. 现代情报, 2018, 38(1): 92- 99. (HE J, LIU Y Z, WANG J K. Community classification and group recommendation strategy of online social networks based on social selection and social influence[J]. Journal of modern information, 2018, 38(1): 92-99.)
[18] CHEN Y L, CHUANG C H, CHIU Y T. Community detection based on social interactions in a social network[J]. Journal of the Association for Information Science & Technology, 2014, 65(3): 539-550.
[19] 蒋璐, 陈云伟. 多节点多关系的混合网络社团划分研究综述[J]. 图书情报工作, 2021, 65(19): 142-150. (JIANG L, CHEN Y W. A review of community detection in hybrid networks with multiple nodes and multiple relationships[J]. Library and information service, 2021, 65(19): 142-150.)
[20] HUTAIR M B, AGHBARI Z A, KAMEL I. Social community detection based on node distance and interest[C]//Proceedings of the 3rd IEEE/ACM international conference on big data computing, applications and technologies. Shanghai: IEEE, 2016: 274-289.
[21] DANG T A, VIENNET E. Community detection based on structural and attribute similarities[J]. Achi, 2012, 659: 7-12.
[22] 金红, 胡智群. 基于非负矩阵分解的稀疏网络社区发现算法[J]. 电子学报, 2023, 51(10): 2950-2959. (JIN H, HU Z Q. The non-negative matrix factorization based algorithm for community detection in sparse networks[J]. Acta electronica sinica, 2023(10): 2023, 51(10): 2950-2959.)
[23] 郑裕龙, 陈启买, 贺超波, 等. 图卷积网络增强的非负矩阵分解社区发现方法[J]. 计算机工程与应用, 2022, 58(11): 73-83. (ZHENG Y L, CHEN Q M, HE C B, et al. Nonnegative matrix factorization community detection method enhanced by graph convolutional network[J]. Computer engineering and applications, 2022, 58(11): 73-83.)
[24] 林晓炜, 陈黎飞. 结构扩展的非负矩阵分解社区发现算法[J]. 山东大学学报( 工学版), 2021, 51(2): 57-64, 73. (LIN X W, CHEN L F. Community detection using nonnegative matrix factorization with structure extension [J]. Journal of Shandong University (engineering science), 2021, 51(2): 57-64, 73.)
[25] 刘峰, 佐斌. 群际情绪理论及其研究[J]. 心理科学进展, 2010, 18(6): 940-947. (LIU F, ZUO B. Intergroup emotions theory and the researches[J]. Advances in psychological science, 2010, 18(6): 940-947.)
[26] MACKIE D M, SMITH E R, BANERJI I, et al. Group-based emotion processes generalize across group exemplars and types[J]. Group processes & intergroup relations, 2024, 27(4): 925-945.
[27] WAKEFIELD R L, WAKEFIELD K. The antecedents and consequences of intergroup affective polarisation on social media[J]. Information systems journal, 2023, 33(3): 640-668.
[28] FIGUEIREDO A, VALENTIM J P, DOOSJE B. Theories on intergroup relations and emotions: a theoretical overview[J]. Psychologica, 2014, 57(2): 7-33.
[29] GOLDENBERG A, GARCIA D, SURI G, et al. Collective emotions[J]. Current directions in psychological science, 2020, 29(2): 154-160.
[30] 张梦瑶, 朱广丽, 张顺香, 等. 基于情感分析的微博热点话题用户群体划分模型[J]. 数据分析与知识发现, 2021, 5(2): 43-49. (ZHANG M Y, ZHU G L, ZHANG S X, et al. Grouping microblog users of trending topics based on sentiment analysis[J]. Data analysis and knowledge discovery, 2021, 5(2): 43-49.)
[31] NGUYEN T, PHUNG D, ADAMS B, et al. A sentimentaware approach to community formation in social media[C]//Proceedings of the international AAAI conference on Web and social media. Oxford: AAAI, 2012, 6(1): 527-530.
[32] 李威. 基于社交网络分析的用户群情感倾向研究[D]. 武汉 : 华中师范大学, 2019. (LI W. Research on the emotional tendency of user groups based on social network analysis[D]. Wuhan: Central China Normal University, 2019.)
[33] MACKIE D M, SMITH E R, BANERJI I, et al. Group-based emotion processes generalize across group exemplars and types[J]. Group processes & intergroup relations, 2024, 27(4): 925-945.
[34] 蔡骐, 翁丽玲. 情感视域下的阅读社群研究[J]. 中国编辑, 2020(11): 16-20. (CAI Q, WENG L L. Research on reading communities from the perspective of sentiment[J]. Chinese editorials, 2020(11): 16-20.)
[35] 褚传弘. 身体漫游、朗读仪式与文化社群——新型城市读书会的地方情感研究[J]. 阅江学刊, 2021, 13(2): 108-118, 123. (CHU C H. Body roaming, reading ceremony and cultural community: research on the local emotion of the new urban reading club[J]. Yuejiang academic journal, 2021, 13(2): 108- 118, 123.)
[36] 王昊懿. 群体情绪分类方法的研究与实现[D]. 西安: 西安电子科技大学, 2021. (WANG H Y. Research and implementation of group-level emotion classification method[D]. Xian: Xidian University, 2021.)
[37] 高帅鹏, 王怡凡. 基于图像的群体情绪识别综述[J]. 计算机与现代化, 2024(8): 98-107. (GAO S P, WANG Y F. Survey on group-level emotion recognition in images. computer and modernization, 2024(8): 98-107.)
[38] 杜世强, 石玉清, 王维兰, 等. 基于图正则化的半监督非负矩阵分解[J]. 计算机工程与应用, 2012, 48(36): 194-200. (DU S Q, SHI Y Q, WANG W L, et al. Graph regularized-based semi-supervised non-negative matrix factorization[J]. Computer engineering and applications, 2012, 48(36): 194-200.)
[39] TRIGEORGIS G, BOUSMALIS K, ZAFEIRIOU S, et al. A deep matrix factorization method for learning attribute representations[J]. IEEE transactions on pattern analysis and machine intelligence, 2016, 39(3): 417-429.
[40] DENG C, HE XF, HAN JW, et al. 2011. Graph regularized nonnegative matrix factorization for data representation. [J]. IEEE transactions on pattern analysis and machine intelligence, 2010, 33(8): 1548-1560.
[41] MACQUEEN J. Some methods of classification and analysis of multivariate observations[C]//Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, volume I: statistics. Berkeley: University of California Press, 1967: 281- 297.
[42] ESTER M, KRIEGEL H P, SANDER J, et al. A density-based algorithm for discovering clusters in large spatial databases with noise[C]//kdd. 1996, 96(34): 226-231.
[43] NG A Y, JORDAN M I, WEISS Y. On spectral clustering: analysis and an algorithm [J]. Advances in neural information processing systems, 2001, 14: 849-856.
[44] 高大菊. 基于计算机大数据分析的社交媒体用户行为挖掘与情感分析研究[J]. 信息记录材料, 2024, 25(10): 118- 120. (GAO D J. Research on social media user behavior mining and sentiment analysis based on computer big data analysis[J]. Information recording materials, 2024, 25(10): 118-120.)
[45] JIANG J, WANG L. A study of the influence of different opinion leaders on the evolution of followers’ views in social e-commerce networks[C]//202120th international symposium on distributed computing and applications for business engineering and science. Nanning: IEEE, 2021: 49-52.
Outlines

/

〈 〉