Using Deep Learning Model to Predict Terms Use by Terrorist to Pre-Plan an Attack on A Real-Time Twitter Tweets from Rapid Miner

Terrorism has an impact on people's quality of life by stifling their physical and mental freedom. People are researching harder to investigate new defense systems to defend humanity. To protect people's lives in society and to raise the overall standard of living, counterterrorism strategies have been employed. Recently, artificially intelligent tools for counterterrorism have been developed using machine learning methodologies (AI). In this research, deep learning approaches are researched to comprehend the behavior of terrorist operations because this field of machine learning has recently seen a rise in its popularity. The first experiment, consisting of 10,000 tweets, was collected using five different terrorist terms, and the second experiment, 6,000 tweets from different features, was collected using Rapid Miner and a spreadsheet to collect real-time data. The models are implemented in deep learning model and two machine learning algorithms, i.e. Naïve bayes and kNN using Rapid Miner. The performance of the deep learning model is compared with two machine learning algorithms, and it is shown that the performance of deep learning is 79 to 89 percent in terms of accuracy, precision, recall, and Fl-Score, while traditional machine learning algorithms have only been able to achieve an accuracy of up to 68 percent. This leads to the conclusion that deep learning is an adequate model to utilize for forecasting terrorist activity. As shown by our studies, the dataset for terrorist activities is enormous. A deep learning model is a good choice to handle large datasets and comprehend the underlying patterns in the dataset.