Comments · 47 Views

Global Terrorism is an escalating global agitation over the years despite the development. Since the attacks of 9/11, the structure of terrorism has seen tremendous extension and acquired attention from divergent communities worldwide. Global Terrorism Database (GTD) includes terrorist incidents that have occurred from 1970 through 2015 and now includes more than 180,000 attacks. With details on various dimensions of each attack, the GTD familiarizes analysts, policymakers, scholars, and journalists with patterns of terrorism. This paper examines the Global Terrorism Database (GTD) yielding its analysis and prediction repercussions using data mining tools.


All intelligence-gathering organizations face a massive challenge in effectively analyzing increasing volumes of crime data. Data Warehouse and mining is a powerful tool that enables us to explore large databases quickly and efficiently. This paper examines how data are selected and categorized in the GTD, drawing attention to changing methodological practices since the 1970s. This article, using GTD, highlights the geographic distribution and intensity of terrorist attacks, the sources and tactics of terrorism, as well as its effects on people and property. The growing potential of data analytical tools like AWS’s Quicksight, Tableau, MS Access and MS Excel are essential in discovering knowledge about the structure of terrorist organizations which is important for developing effective combating strategies against terrorism. Given the data, this article illustrates the tremendous impact of these terrorist attacks on different geographical regions, analysis on the prime time of these occurrences and further analysis was concentrated on the type of attack that were successful. Countries such as Afghanistan, Indonesia, India, and Iraq that have been severely impacted by terrorism have been analyzed on what time of the year they were attacked the most.


The war on terror has taken center stage since the 9/11 attacks. Now, information-related issues, such as the communication and sharing of research ideas among counterterrorism researchers and the dissemination of counterterrorism knowledge among the general public, become critical in detecting, preventing, and responding to terrorism threats. This Global Terrorism Database (GTD) article has consolidated both domestic and international terrorist activities between 1970 and 2016. With the voluminous data represented by the GTD, the challenge now becomes to understand and decipher important patterns and relationships. With advanced tools today, the article helps identify trends and strategic analysis of high-level patterns and tactical analysis of individual events.


Problem Statement

Problem 1: Which attack approach has affected the highly terrorized regions?

Terrorism is often regionally-focused. It’s also highly concentrated within specific countries. In this chart, we see the number of deaths from terrorism by region and attack type. Major events that most people would understand to be terrorism are roadside bombings, attacks on religious, car detonations or political institutions. Of the overall global deaths from terrorism included in the Global Terrorism Database, 92% occurred in the Middle East (Iraq, Afghanistan), Pakistan or India. Less than 8% of deaths were in Europe, the Americas and Oceania combined. This is also true when we look at the number of incidents, rather than the number of deaths. There is a strong regional focus but this is also heavily concentrated in only a few countries within these regions.  Most victims of terrorism die in the Middle East, Africa and South Asia. This hasn’t always been the case. Guerrilla movements in Central and South America, for example, dominated terrorism in the 1980s. The chart below also illustrates that bombing was the most used attack type on the severely affected regions.

Fig: The chart above shows different attack approaches with rate of success against affected countries


Problem 2: Has the volume of terrorist attacks increased over the years?

It can be difficult to separate a rise in attention from a rise in frequency. Increasing attention on terrorism can therefore make it seem like it’s always getting worse. But is this really true? The visualization below shows the year with the maximum attacks in each decade of the 70s, 80s, 90s, 2000s and 2010s. Data from the most comprehensive database to date - the Global Terrorism Database was used for the analysis. The chart shows the annual deaths from terrorism in the order of hundreds, and reaching over 400 deaths in some years. Based on fatalities we see terrorism was relatively high in the 1970s, then comparably ‘quiet’ – with exception of major outlying years, 1989, 1991 and 2001 – in the decades which followed. The year to year changes are nonetheless volatile. When we look at the number of terrorist attacks we see a marked decline since the early 1970s.


Fig: The chart above shows frequency of attacks with rate of success over each decade.


Problem 3: Are Terrorist Attacks More Likely to Occur on Key Dates?

Warnings of possible terrorist attacks and security are often increased on key dates, which may have symbolic importance to the terrorists or, in their eyes, are important to their foes—for example, July 4 (Independence Day of the United States) or the anniversary of the September 11 twin tower attacks. Certainly, both terrorists and those charged with security think more about these dates, but do more terrorist attacks, in fact, occur on them? To answer this question, the chart used from the distribution of terrorist attacks from 1970 to 2015 by quarter is checked to see if certain quarters of a year stood out. The attacks were prominent around the months of April, May, and June (Quarter 2). On each of these months of the quarter, communities hold events that draw large crowds, which raises the threat of terrorism by creating potential soft targets. Furthermore, any event that attracts a large crowd requires increased security for public safety. More particularly, specific dates that one might think would bring an amplified risk of terrorism do not appear associated with a bigger number of events historically. For example, the U.S. data did not include any fatal terrorism events on September 11 following 2001, although we know that terrorists contemplated attacks on these days. Similarly, none of the dates during the observance of Ramadan has experienced significantly more terrorism globally than any other dates.

Fig: The chart above shows possible terror attacks with rate of success over quarters with key dates.


Strategic Alternatives

In the era of big data, the advent of latest technologies, the use of social media, mobile technologies, and cloud drives has led to the generation of different types of voluminous data in the form of big data. This provides wider attention to strategic alternatives in tackling global terrorism and preventing its infiltration in the society.

Detecting suspicious data using Big Data to counter Global Terrorism:

 The use of big data analytical tools and techniques is very useful for identifying sensitive data, because of the usage of parallel and distributed computing to process voluminous data. Top security agents will always looks for ways to be one step ahead of terrorists to prevent unexpected incidents. Hadoop clusters helps determine precise data from huge varieties of data to support Social Justice Organizations in combating terrorist activities on a global scale. To achieve this goal, algorithmic approaches which are part of NLP (Natural Language Processing) like parallelization, annotations, lemmatization, stop word Remover, term frequency and inverse document frequency, and singular value decomposition, can be successfully implemented.

Spark, Resilient Distributed Datasets (RDD) is used to achieve parallelism and work distribution. Moreover, files will be read using Spark API. Work efficacy is achieved through parallelism, Spark Context Driver, Cluster Manager, Worker Nodes were communicated with each other. Annotation is a process in which raw data is passed to the Annotation Object, and after processing with the relevant function, an annotated text is generated. This technique is used to find patterns and inferences in the datasets. As the next step, Lemmatization is an NLP process where the base or dictionary form of a word from plain text is acquired. Even after using these techniques, there might be undesired terms and information in the analysis. In order to reduce the dimensional space, Spark API I with StopWord Remover functions can be implemented

Tracking terrorist social networks using visualization techniques

Social network analysis techniques are helpful in the visual inspection of the network global structure. Two interactive visualization techniques for tracing complex terrorist social networks are: fisheye views and fractal views. These techniques facilitate enhanced view of regions of interest by allowing a user to select one or more focus points and dynamically adjusting the graphical layout. Previously unreadable patterns in the normal display can be effectively recognized by an investigator with the combination of the above mentioned techniques.

Generally, social network analysis is mainly a manual process. Database searches and reading reports to find useful entities and relationships in a large network is time-consuming and labor intensive. Xu and Che adopted the metric multidimensional scaling algorithm to visualize the criminal social networks. The structure of relatively small and simple networks can be inferred from static graphical layout. However, it is usually not efficient for the manual exploration of large and complex networks. Two interactive visualization techniques such as fisheye views and fractal views is used for facilitating the analysis of complex social networks and demonstrate its application in the analysis of large terrorist networks.

Using the 5W investigative analytical approach to combat global terrorism

For better understanding of terrorist activities by investigators, a visual analytical system is used that focuses on the five W’s (who, what, where, when, and why). Views in our system are highly correlated, and each represents one of the W’s. This approach enables an investigator to interactively explore terrorist activities efficiently and discover reasons of attacks (why) by identifying patterns temporally (when), geo-spatially (where), between multiple terrorist groups (who), and across different methods or modes of attacks (what). The collective information gathered from asking these five questions allows analysts to think tactically and strategically. This technique facilitates communication of investigative findings and hypotheses among analysts. Temporal trends in relation to both geographical and other patterns are depicted by the model which also suggests possible future direction of events. It can also reduce ambiguity and effort in communicating the results of analysis.



The GTD – as with other terrorism databases – are curated through records and analysis of print and electronic media. The collation of incidents across the world today and in the recent past is sufficiently complete to understand the global distribution of terrorist incidents and how they have changed over time. This article illustrates the understanding of the sources and frequency of terrorism which has a significant impact on many areas of society and policy, including immigration, counterterrorism efforts, and international relations. There is a strong regional focus but this is also heavily concentrated in only a few countries within these regions.  Most victims of terrorism die in the Middle East, Africa and South Asia. The frequency of attacks was prominent in the 1970s. The relative “tranquility” during the years since 9/11 compared with the more turbulent 1970s may be explained by the following factors. In the 1970s, there were more terrorist groups operating with definable constituencies than there are today. When we look at the number of terrorist attacks we see a marked decline since the early 1970s. The absence of increases in terrorism on significant dates indicates that specific dates that one might think would bring a heightened risk of terrorism do not appear associated with a greater number of events historically. Data Analysis and development of tools in the analytics domain helps in designing strategies to minimalize such violent threats in the near future.


Reid, J. Qin, W. Chung, J. Xu, Y. Zhou, R. Schumacher, et al., "Terrorism knowledge discovery paper: A knowledge discovery approach to addressing the threats of terrorism", International Conference on Intelligence and Security Informatics, pp. 125-145, 2004, June.

A H Wahbeh, Q A Al-Radaideh et al., "A comparison study between data mining tools over some classification methods", International Journal of Advanced Computer Science and Applications Special Issue, pp. 18-26, 2011.

Analyzing the Terrorist Social Networks with Visualization Tools -

Yamasue, K. Kasai, A. Iwanami, T. Ohtani, H. Yamada, O. Abe, et al., "Voxel-based analysis of MRI reveals anteriorcingulate gray-matter volume reduction inposttraumatic stress disorder due to terrorism".

Crime data mining: a general framework and some examples -

The Global Terrorism Database (GTD): Accomplishments and Challenges

Investigative Visual Analysis of Global Terrorism

Terrorism Knowledge Discovery Project: A Knowledge Discovery Approach to Addressing the Threats of Terrorism -