Читать книгу Intelligent Data Analytics for Terror Threat Prediction - Группа авторов - Страница 12

Ravi Kishore Devarapalli* and Anupam Biswas†

Department of Computer Science Engineering, National Institute of Technology Silchar, Assam, India

Abstract

Social media like Facebook, Twitter, WhatsApp, Sina Wiebo, Hike, etc. play an important role in the information spreading in today’s world. Due to large scale connectivity, some of the cyber criminals are choosing these platforms to implement their criminal activities such as rumor spreading which is popularly known as rumor diffusion. In a pluralistic society like India, rumors that spread over various social networking platforms are much more vulnerable. It is challenging for even public, government and technical experts working on social media to find these cyber-crimes and its origin to punish the culprit. Nowadays, some cyber criminals are choosing different platforms and paths to accomplish their plans to escape brilliantly from those activities. Thus automated detection of rumor and tracing its source has great importance.

This chapter surveys the different automated rumor detection systems in social networks and techniques to trace the source of rumor. Detection of rumors on social networking platforms is possible through analysis of shared posts and comments wrote by followers. After detection of rumors, next job is to 1) prevent the rumors from further spreading and 2) identification of culprit i.e. the originator of rumor. This chapter covers both these aspects. The opinion of an influential person in the group influences others very easily. Cyber criminals may use separate communities in social media to fulfil their activities. Thus, it is important to trace the most influential person in community to prevent further spreading. This chapter aims to discuss the recent techniques that are developed for identifying influential persons in the group. The chapter also aims to study the various techniques developed for identifying the culprit, which are based on factors like network structure, diffusion models, centrality measures, etc. The chapter will also discuss the various challenges including real-life implementation, evaluation, and datasets in respect to both rumor detection and rumor source tracing.

Keywords: Rumor, rumor source detection, centrality measures, social networks, diffusion models