KDD 2017 Tutorial: Data-Driven Approaches towards Malicious Behavior Modeling

Œe safety, reliability and usability of web platforms are o‰en compromised by malicious entities, such as vandals on Wikipedia, bot connections on TwiŠer, fake likes on Facebook, and several more. Computational models developed with large-scale real-world behavioral data have shown signi€cant progress in identifying these malicious entities. Œis tutorial discusses three broad directions of state-of-the-art data-driven methods to model malicious behavior: (i) feature-based algorithms, in which distinguishing behavioral features are proposed to predict the malicious users; (ii) spectralbased algorithms, which have been widely used in seŠings of directed graphs, undirected graphs, and bipartite graphs such as “who-follows-whom” TwiŠer data and “who-likes-what” Facebook data; and (iii) density-based algorithms, which eciently look for suspicious, highly-dense components in multi-dimensional behavioral data. Œis tutorial will introduce the details of the general algorithms from the above three classes that can be applied to any platform and dataset. ACM Reference format: Meng Jiang, Srijan Kumar, V.S. Subrahmanian, and Christos Faloutsos. 2017. KDD 2017 Tutorial: Data-Driven Approaches towards Malicious Behavior Modeling. In Proceedings of ACM SIGKDD conference, Halifax, Nova Scotia, Canada, August 2017 (SIGKDD’17), 4 pages.