Detecting bots in Internet chat
MetadataShow full item record
Internet chat is a real-time communication tool that allows on-line users to communicate via text in virtual spaces, called chat rooms or channels. The abuse of Internet chat by bots also known as chat bots/chatterbots poses a serious threat to the users and quality of service. Chat bots target popular chat networks to distribute spam and malware. We first collect data from a large commercial chat network and then conduct a series of analysis. While analyzing the data, different patterns were detected which represented different bot behaviors. Based on the analysis on the dataset, we proposed a classification system with three main components (1) contentbased classifiers (2) machine learning classifier (3) communicator. All three components of the system complement each other in detecting bots. Evaluation of the system has shown some measured success in detecting bots in both log-based dataset and in live chat rooms.