Cerberus: cross-site social bot detection system based on deep learning

Abstract

Social networking sites have attracted billions of users and influence people's lifestyles. However, as open platform with low requirements for registration and joining, it is inevitable that social bots are able to easily register and do harmful things such as controlling public opinions and spreading inaccurate information for profit. Nevertheless, single-site social bot detection systems often rely on historical behavioral data to identify bots, and the detection occurred after the social bots have implemented their attacks. To identify social bots as early as possible, this paper proposed Cerberus, a cross-site social bot detection system. Cerberus can solve the cold-start problem of user identification caused by insufficient user data on a single platform at an early stage. Cerberus used personal information and historical activity on the Medium website of users to make prediction about whether a user's account on Twitter was a social bot. The results from our experiments show that the AUC score of Cerberus can reach 0.7522, which has good recognition accuracy.

Keywords

• online social network
• Social Bot Detection
• cross-site linking
• deep learning
• cold-start user