A Framework for Predicting and Identifying Radicalization and Civil Unrest Oriented Threats from WhatsApp Group

Social media is an emerging area of research for data scientist. Lots of test dataset are available for Facebook, Twitter, and Instagram. Very few research has been done on WhatsApp group chat log due to scarcity of data openly available. WhatsApp is secured by encryption so WhatsApp is heavily used for terrorist activities, spreading riot, creating civil unrest, and disseminating radicalization. Data to conduct research on these issues are hardly available as these are very sensitive information. We are proposing here a framework to address this type of social issue. By NLP and semi-supervised machine learning, we can design a system which will predict the probability of WhatsApp group to be a threat for civilization. Success of this framework depends on how much actual data we can feed to system. Depending on training data, it will predict with more accuracy. This paper will show how this system can work properly in actual field.