International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 08 | Aug 2024
p-ISSN: 2395-0072
www.irjet.net
Detecting and classifying human terrorist attack type involvement using machine learning Momita Rani Giri1, Prof. Preeti Rai2 1Reseacrh Scholar, Department of CSE, Gyan Ganga Institute of Technology and sciences, Jabalpur, M.P. 2Professor, Departmet of CSE, Gyan Ganga Institute of Technology and Sciences, Jabalpur, M.P.
---------------------------------------------------------------------***-------------------------------------------------------------------to commit crimes against non-combatants [2]. This behavior Abstract – is often designed to intimidate bystanders rather than the victim. Although some people believe that violence has some positive consequences, the truth is that even if it is determined as a bad behavior, it is illegal [3]. It is based on the principle that any behavior that is contrary to the representation of public order is unlawful, will lead to abuse and illegality. Because violence itself includes violence, violence, threats and behavior that endangers public and cultural order [4]. Violence has also been shown to be associated with aggression, violence and conflict [5, 6]. The causes of violence can be divided into three levels: natural events, strategic ideas and personal characteristics. Factors include reasons for action and conditions that allow radicalization and support the ideology of the opposition. In the short term, these activities may mean educational support; but in the long term (in theory), the situation may indicate political, ethnic, oppositional or separatist tendencies. At the same time, it will seek to disrupt and undermine government processes, influence public behavior and good governance, incite fear and sympathy, and provoke protests that are unacceptable to them. Personal characteristics include the criminal's worldview, mental health and personality. Believes there are good and bad people. The threat to life and property posed by international terrorism and the need to control it is a good reason to conduct this research. The impact of machine learning and artificial intelligence in preventing the spread of crime cannot be overstated. This technology can help prevent and combat crime, assist governments and other decision-makers, and coordinate the efforts of citizens and travelers. Preventing terrorism is effective in the context of the type of violence experienced in a region and in terms of protecting the lives and properties of citizens [7-9]. Machine learning can be used to predict crime based on data obtained from financial transactions, travel patterns, events, and media such as social media. The need for this research will reveal the importance of international crime data that can provide important information about attacks [10, 11]. Study statistics to learn how to combat international terrorism. Some authors use machine learning such as naive bayes (nb), nearest neighbor (kNN), decision trees, and support vector machine (svm) to predict the role of the aid group in the events [12, 13]. Some authors also published a proposal to use deep learning to predict the speed of online criminals [14]. Developed a hazard system to identify terrorist attacks
Terrorism can be defined as the use of violence against people or property to threaten or coerce a government or its citizens to achieve certain political or social goals. This is a global problem that results in loss of life and property and negatively affects the economy and international trade. Terrorism is also associated with high uncertainty, and many countries around the world are interested in any research that can reduce the threat of terrorism. Most research on crime focuses on measures to prevent crime or how to reduce crime, but there is limited research on what makes crime evil. The goal of this project is to develop a machine learning model that uses linear support vector machine to predict the likelihood that criminals will engage in activities that support crime. Data from global crime database and data preprocessing including data cleaning and dimension reduction. Then svm machine learning model is built and applied to the selected features. Logistic regression of svm model is better than baseline model. This means svm uses weighted objects as selection method to get the best results among certain options to predict crime scene. Our results show that svm machine learning models can predict malicious activities.
Keywords: Terrorist Activity Recognition, Terrorist Activities, Global Terrorism Database, Logistic Regression, Random Forest, and Linear SVM.
1. INTRODUCTION Terrorism is a global threat that has been with humanity since ancient times. It is a global problem because it causes loss of life, property and security both domestically and internationally. Previous studies have shown that the level of insecurity and uncertainty caused by violence affects decision-making to the extent that more people make more decisions. There is a risk that a single decision can lead to the insecurity that is often associated with terrorism. Violence [1]. One of the worst acts of terrorism, September 11, is not only recorded as one of the deadliest terrorist attacks in history, but the interest in rapidly detecting, predicting, and eliminating enemy activities and efforts, supported around the world, is called terrorism. . . . Article 22 of the United States convention defines terrorism, but the definition of terrorism is often the subject of specific treaties. The license
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 419