An Effective Approach to Terrorist Threat Prediction Using Data Fusion for Warning Signals

Saurabh Singh^1,*, Shashi Kant Verma², Akhilesh Tiwari³

¹Department of Computer Science and Engineering, Jabalpur Engineering College, Jabalpur, Madhya Pradesh 482001, India

²Computer Science Department, Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal, Uttarakhand 246194, India

³Department of CSE & IT, Madhav Institute of Technology and Science, Gwalior, Madhya Pradesh 474005, India

Email: ssingh@jecjabalpur.ac.in; skverma.gbpec@rediffmail.com; atiwari.mits@gmail.com

Abstract

Terrorist network research is crucial for both anticipating terrorist acts and extracting valuable information from existing unauthenticated sources. The best method for deciphering intricate terror networks is graphic analysis. The suggested study used a data fusion technique to analyze the terrorist network using the dataset from the terrorist assault in Mumbai on 26/11. In order to effectively forecast the terror threat, the study also focuses on finding the critical node. Wassi led the attack and was a key controlling agent, according to the measurement analysis. The information matched the government's report.

Keywords: Terrorist Network; Data mining; Threat prediction; Graphics processors; Data models

1.                   Introduction

Terrorism is one of the significant concerns for government agencies and law enforcement entities across the globe. With enormous advancements in perilous technology, the detrimental interest and destructive capability of the noxious terrorist organizations have accelerated tremendously. Vigorously monitoring terrorist organizational networks and tracking their vicious activities is a challenging task for government and intelligence agencies. Significant resource quantity is essential for monitoring these networks [1]. Most of the time, law enforcement agencies and intelligence entities lack technologically advanced resources to track the activities of terrorist organizations effectively. Hence, comprehensive techniques for analyzing structured data and raw data to unmask critical and perilous data are required. These techniques strengthen the tracking and monitoring capability of the intelligence agencies in identifying the threats from terrorist organizations [2]. Data fusion approaches are gaining significance due to their expanded approach to collecting and analyzing multiple data from various sources. Data fusion approaches are concerned with the prediction and evaluation of the current state of one or more suspected entities. Multiple evaluations of target recognition and target tracking computations are assumed while processing multiple data. Other data fusion techniques involve the use of contexts to derive different states for different entities. These contexts can consist of the relation among other entities of interest and context type and appropriate situation [3]. Data fusion approaches exhibit the capability of exploiting the situational and relational contexts to characterize and identify the relationship between various situations. Situation and threat assessment are two major significant data fusion concepts that are dependent on each other. Generally, they are treated jointly in control processes by military commanding authorities. According to Waltz and Llinas, assessment of the situation provides an overview of the suspected areas in terms of the suspicious activities, perilous functionalities, manoeuvres, and organizational aspects of the terrorist groups, and from the obtained overview, one can infer about the forthcoming outcome and be prepared to combat the upcoming threat [4]. While threat assessment approximates the degree of