近年來社群媒體越來越興盛，改變了大眾接收資訊的習慣。社群媒體上的話語權也成了不同陣營攻防的重點。因此社群媒體上開始出現越來越多行為可疑的異常帳號，試圖操弄輿論引導風向，藉此獲得自身的利益。
本論文研究開發一個社群媒體上的異常帳號探勘分析，透過發文及留言互動行為探索分析異常帳號。本系統的核心精神為由已知的異常帳號找出其他未知異常帳號。系統提供兩大主要功能Explore By Examples及Analysis By Examples。前者著重在探索異常帳號、後者著重在分析異常行為。兩大功能都由作息、來源IP、回應行為、網絡四個面向來探索分析，以協助使用者探索出異常帳號。本論文以台灣最熱門的本土社群媒體平台PTT BBS資料進行實證，並根據PTT官方公告異常帳號運用本系統進行案例分析。

In recent years, with the growth of social media, public's habits of receiving information have been changed. More and more malicious accounts appear on social media. These malicious accounts try to manipulate public opinion for spin control to gain their own interests.
This thesis aims at the design and implementation of an intelligent discovery system to explore malicious accounts over social media. The core spirit of the developed system lies in the exploration of unknown malicious accounts by known malicious accounts over social media. Two main functions provided by the developed system are exploration by examples and analysis by examples. While the former focuses on exploration of unknown malicious accounts, the latter focuses on the analysis of known ones. To explore and analyze malicious account, four aspects of malicious behaviors are considered, namely activity, IP address, like/dislike behaviors, and network. The developed system is verified on the data collected from PTT bulletin board system, which is the most popular domestic social media service in Taiwan. Four case studies are performed to demonstrate the superiority of the developed intelligent discovery system.

[1] K. S. Adewole, N. B. Anuar, A. Kamsin, K. D. Varathan and S. A. Razak, Malicious Accounts: Dark of the Social Networks, Journal of Network and Computer Applications, 79, pp. 41-67, 2016.
[2] A. Badawy, E. Ferrara and K. Lerman, Analyzing the Digital Traces of Political Manipulation: The 2016 Russian Interference Twitter Campaign, The 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, pp. 258-265, 2018.
[3] N. Chavoshi, H. Hamooni and A. Mueen, Temporal Patterns in Bot Activities, International Conference on World Wide Web Companion, pp. 1601-1606, 2017.
[4] N. Chavoshi, H. Hamooni, and A. Mueen, DeBot: Twitter Bot Detection via Warped Correlation, IEEE International Conference on Data Mining, pp. 817-822, 2016.
[5] N. Chavoshi, H. Hamooni, and A. Mueen, Identifying Correlated Bots in Twitter, International Conference on Social Informatics, pp. 14-21, 2016.
[6] L. A. Cornelissen, R. J. Barnett, P. Schoonwinkel, B. D. Eichstadt and H. B. Magodla, A Network Topology Approach to Bot Classification, Annual Conference of the South African Institute of Computer Scientists and Information Technologists, pp. 79-88, 2018.
[7] C. A. Davis, O. Varol, E. Ferrara, A. Flammini and F. Menczer, BotOrNot: A System to Evaluate Social Bots, International Conference Companion on World Wide Web, pp. 273-274, 2016.
[8] A. Duh, M. S. Rupnik, and D. Korošak, Collective Behavior of Social Bots Is Encoded in Their Temporal Twitter Activity, Big Data 6, 2, pp. 113-123, 2018.
[9] S. Gokalp, M. Temkit, H. Davulcu, and I.H. Toroslu, Partitioning and Scaling Signed Bipartite Graphs for Polarized Political Blogosphere, IEEE International Conference on Social Computing, pp. 168-173, 2013.
[10] S. Gupta, P. Kumaraguru, and T. Chakraborty, MalReG: Detecting and Analyzing Malicious Retweeter Groups, ACM India Joint International Conference on Data Science and Management of Data, pp. 61-69, 2019.
[11] M. Jiang, P. Cui and C. Faloutsos, Suspicious Behavior Detection: Current Trends and Future Directions, IEEE Intelligent Systems, 31, 1, pp. 31-39, 2016.
[12] A. Karataş and S. Şahin, A Review on Social Bot Detection Techniques and Research Directions, International Security and Cryptology Conference, Turkey, pp. 156-161, 2017.
[13] T. Khaund, K. K. Bandeli, M. N. Hussain, A. Obadimu, S. Al-Khateeb and N. Agarwal, Analyzing Social and Communication Network Structures of Social Bots and Humans, The 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, pp. 794-797, 2018.
[14] S. Kudugunta and E. Ferrara, Deep Neural Networks for Bot Detection, Information Sciences, 467, pp. 312-322, 2018.
[15] T. Mikolov, K. Chen, G. Corrado, and J. Dean, Efficient estimation of word representations in vector space, International Conference on Learning Representations, abs/1301.3781, 2013.
[16] S. Sadiq, Y. Yan, A. Taylor, M. L. Shyu, S. C. Chen and D. Feaster, AAFA: Associative Affinity Factor Analysis for Bot Detection and Stance Classification in Twitter, 2017 IEEE International Conference on Information Reuse and Integration, pp. 356-365, 2017.
[17] H. Sakoe and S. Chiba, Dynamic programming algorithm optimization for spoken word recognition, IEEE Trans Acoustics Speech Signal Process, 26, pp. 43-49, 1978.
[18] R. Schuchard, A. Crooks, A. Stefanidis and A. Croitoru, Bots in Nets: Empirical Comparative Analysis of Bot Evidence in Social Networks, L.M. Aiello, C. Cherifi, H. Cherifi et al.(eds) Complex networks and their applications VII. Springer International Publishing, pp. 424-436, 2019.
[19] Y. Wang, C. Wu, K. Zheng, and X. Wang, Social Bot Detection Using Tweets Similarity, International Conference on Security and Privacy in Communication Systems, pp. 63-78, 2018.
[20] 黃懷萱，利用行為脈絡探索社群媒體上的異常使用者，國立政治大學資訊科學系碩士論文，2020。