Our recent work on detecting automated fake account has been published at Elsevier Journal on Computers and Security. This article has a more comprehensive treatment as compared to the conference paper, and includes supplementary evaluation results on new datasets and the formal analysis of Íntegro: Our account ranking system for aiding OSNs in detecting malicious, automated fake accounts (a.k.a. socialbots).
It’s my pleasure to announce that I’ll be giving a hands-on tutorial on social web intelligence in February 21st at UBC.
Organized by the SFU/UBC Salon Series on Digital Social Sciences, Humanities and the Arts, the tutorial/workshop will introduce the theory and practice of basic concepts in network analysis, machine learning, and data mining to make sense of the social and information networks that have been fuelled and rendered accessible by the Internet.
Participants will learn about the structure and evolution of networks, drawing on knowledge from disciplines as diverse as sociology, mathematics, statistics, computer science, economics, and physics.
Interactive demonstrations and hands-on analysis of real-world data sets will focus on a range of tasks: from online network data collection, to identifying important nodes in the network, to detecting communities, to opinion mining and sentiment analysis, to predicting future relationships and social attributes.
Telefonica has just open-sourced Grafos ML, a system and tools for large-scale machine learning and graph analytics on top of Apache Giraph. The two main components are Okapi ML Library (machine learning algorithms, which include our recent work on detecting fake accounts in online social services) and RT-Giraph (incremental processing on top of Giraph). This is still an active project that is under heavy development at Telefonica Research, Barcelona.