Detection of terrorism threat elements with natural language processing

In today's world, it has become important to make millions of data understandable. In order to take faster steps in criminal matters, especially by using these data, data analysis should be done quickly. In this context, sentiment analysis performed with the natural language processing method of artificial intelligence enables the elimination of possible loss of life and property. In addition, by listening to all radio frequencies at the same time in possible terrorist areas, the attacks of terrorist organizations can be analyzed with natural language processing methods, so that the attack can be prevented before it takes place. In this master's thesis study, natural language processing methods of artificial intelligence were used in the analysis of text, audio, and image data in the virtual environment for the detection of terrorist threat elements. In this way, it is aimed to ensure the healthy intervention of law enforcement officers and the security of life by analyzing the talks of terrorist elements in terror zones. In the thesis study, the conceptual structure and natural language processing methods related to natural language processing are mentioned in detail. Models were created with the mentioned methods and the results were presented.