Abstract: - Now a day’s social platform arebecoming an main part of life. Almost everyone with smart phone using onlinesocial network. Terrorist organizations use different social media as a toolfor spreading their views and influence general people to join their terroristactivities. Twitter is the most common and easy way to reach mass people withina small amount of time. In this paper, we have focused on the development of asystem that can automatically detect terrorism-supporting tweets by real-timeanalyzation. In this system, we have developed a frontend for real-time viewingof the tweets that are detected using this system. We have also compared theperformance of two different machine learning classifiers, Support VectorMachine (SVM) and Multinomial Logistic Regression and found the first one worksbetter. As our system is highly dependent on data, for more accuracy we added are-train module. By using this module wrongly classified tweets can be added tothe training dataset and train the whole system again for better performance.This system will help to ban the terrorist accounts from twitter so that theycan’t promote their views or spread fear among general people.
KEYWORDS: Terrorism, Social Network, Attacks, Performance Analysis