Antisocial behavior in online environment is one of the most actual and serious problems, which significantly threatens not only the principles, on which the web was built, but also has a critical overreach to society. As a typical example, we can take the spread of misinformation through social networks, which influences opinions of people and consequently, their decisions (e.g. during voting or taking a medical treatment). Another typical example is spread of hate speech, trolling or cyber-bullying in online space. These negative situations have been unintentionally enabled by the rise of information technologies.
Now information technologies may become the means for better understanding and dealing with this phenomenon. The successful reaching of this goal requires a multidisciplinary research, which we address by the project REBELION. The primary focus of this project is the research of models and methods for automatic identification of antisocial behavior. We build explicit models based on analysis of integrated datasets, but also on qualitative understanding of malicious content and behavior. The recognition methods are based on these models and will employ machine learning, advanced automated analysis of natural language texts and user modeling (either of content author or consumer).
The main goal of the project is to research new knowledge in informatics and information technologies, especially: