SocialBot: Behavioral Analysis and Detection

• Published in: Security in Computing and Communications Vol. 625; pp. 450 - 460
• Main Authors: Dewangan, Madhuri, Kaushal, Rishabh
• Format: Book Chapter
• Language: English
• Published: Singapore Springer Singapore Pte. Limited 2016; Springer Singapore
• Series: Communications in Computer and Information Science
• Subjects: Algorithms & data structures; Computer security; Data encryption; Feature extraction; Machine learning; Online Social Network; Social engineering; SocialBots
• ISBN: 9789811027376; 9811027374
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-981-10-2738-3_39

Bots refer to automated software that have the capability to execute commands on receiving instructions from BotMaster. SocialBots are the bots present in Online Social Network (OSN) which mimic the activities of the real users. They have the capability to automatically perform the basic functionalities offered by the OSN platforms. These socialbots have widespread usage in political campaigning and product marketing, but SocialBots can also been used for the purpose of swaying voters, mounting political attacks, manipulating public opinion, etc. Apart from these, SocialBots posses various security risks, one of which is befriending an OSN user thereby gaining access to personal details such as birthday, email id, phone number, address, etc. Detection of these SocialBots is therefore an important problem to be solved in order to maintain the reputation of OSN. Our work concentrates on behavioral analysis of these SocialBots in the OSN and identifying features to be used to develop a model for detection of these Socialbots using machine learning. The model, thus developed, is further used as a background process to create a web-based tool for detection of SocialBots. In our work, we created a SocialBot to perform behavioral analysis. This SocialBot got a good response from the real users and was able to grab 100+ real followers along with some real interactions in form of retweet, mention and direct messages.