透過您的圖書館登入
IP:3.144.97.189
  • 學位論文

這是業配文嗎?網軍的貼文型態與網絡分析

Is This a Fake Review? Analytics on Types of Opinion Posting and Social Network of Spammers

指導教授 : 汪志堅 戴敏育
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


現今社會中,有許多網路寫手存在於各大評論網站,並且這個議題越來越受到許多人的重視。而過去研究中,大多數研究僅針對網路寫手所貼出的文章以及他們的線索進行分析,鮮少根據網路寫手與其它寫手之間會具有一定的關連性進行分析,因此本研究試圖利用社會網絡分析進行偵測網軍。   本研究採用三個實證研究進行偵測網軍,實證研究一發現,以人工的方式無效,但耗費太多的時間、精力及金錢,因此採用自動化的方式進行偵測。實證研究二中採用社會網絡分析特徵值進行分析,並試圖偵測網軍。實證研究三中,則利用社會網絡視覺化的方式進行偵測,將所有作者進行分群,以找到網軍。研究結果顯示,社會網絡分析是可以利用網軍之間的緊密關係找尋網軍。

並列摘要


In recent years, there has many spammer exist in many review websites, and more importantly, the more researchers focus on these issues. However, the past researches focused on the articles spammer wrote and used these articles as a cue to detect them. They few research the relationships among spammers. Thus, this study try the possibility of using social network techniques to detect spammer. This study tried three empirical researches to detect spammers. The result of study 1 indicate that there has no possible to detect by humankind, and it must take a lot of time and money to detect them. Thus, we used an automatic method to detect spammers. The result of study 2, we used social network analysis as features to do it. The result of study 3, we attempt to adopt a real case to analyze the social network of spammers by K-core and Clique analysis. Our research results show that the social connection among spammers is stronger than that among non-spammers. Finally, we propose a novel method of using social network analysis indices as features to detect spammer. The relationship of authors of product reviews posts and their replies might be used for spammer group detection. Moreover, K-cores and cliques can be used as cues to identify spammers.

參考文獻


Mukherjee, A., Liu, B., & Glance, N. (2012). Spotting fake reviewer groups in consumer reviews. Paper presented at the Proceedings of the 21st international conference on World Wide Web.
Chen, Y.-R., & Chen, H.-H. (2015). Opinion spam detection in web forum: a real case study. Paper presented at the Proceedings of the 24th International Conference on World Wide Web.
Borgatti, S. P., Everett, M.G. and Freeman, L.C. (2002). Ucinet for Windows: Software for Social Network Analysis.
DeBarr, D., & Wechsler, H. (2010). Using social network analysis for spam detection Advances in Social Computing (pp. 62-69): Springer.
Freeman, L. C. (1978). Centrality in social networks conceptual clarification. Social networks, 1(3), 215-239.

延伸閱讀