DEVELOPMENT OF TERRORIST THREAT MODEL IN INDONESIA BY USING BAYESIAN NETWORK

• Main Authors: , Hilya Mudrika Arini, , Nur Aini Masruroh, S.T., M.Sc., Ph.D.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/126136/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=66332
• Institution: Universitas Gadjah Mada

Nowadays, many terrorism actions occur around the world. For instance, more convincing evidence of terrorist threat in Southeast Asia came with the devastating attacks on Bali, Indonesia on October 12. From the Bali Bombing, by roughly 202 people died and 209 people were injured. On the other hand, with respect to historical data, there are more than 20 terrorist threats during 2002 to 2012 in Indonesia. According to the massive number of violance due to the threat and the massive number of terrorist threat in Indonesia, preventive solution to eliminate terrorist threat should be applied. However, the preventive solution through studies in national security in Indonesia does not be conducted. Therefore, this study aims to provide preventive solution by developing model of terrorist threat in Indonesia by using Bayesian network. Generally, there are six stages to build the model, started from literature review to what-if scenario. Firstly, from literature review, the initial bayesian belief network can be made. Secondly, the initial bayesian belief network must be verified by the experts.This study uses four experts from researcher, Non Governmental Organization (NGO) in national security system, Indonesian government agencies known as Badan Nasional Penanggulangan Terorisme and the former of Jamaah Islamiyah leader. Thirdly, after bayesian belief network is built, joint probability in each variable is obtained from expert judgment. However, for news and keywords variable, joint probability is obtained from text mining method by using Weka. Fourthly, model testing and validating are conducted by using Genie 2.0. Model is developed by using information before Solo bombing in 25th September 2011, while information before Cirebon bombing in 15th April 2011, Poso bombing in 27th August 2012 and Poso shooting in 3th June 2013 are used for validating the model. Fifthly, to find the most significant variables influencing terrorist threat, sensitivity analysis is conducted by using Netica 1.12. Ultimately, what-if scenario is utilized to find the best scenario to provide the preventive solution for minimizing the occurance of terrorist threat. This study finds several significant findings. Firstly, there are eleven variables which can affect terrorist threat in Indonesia. Secondly, news and the readiness of terrorist group (funding, teamwork and jihad skill) are the most influencing factor which affect terrorist threat in Indonesia. Thirdly, according to several scenarios of the news portion, it can be concluded that the higher positive portion of news, the higher probability terrorist threat will occur. Therefore, the preventive solution to reduce terrorist threat in Indonesia based on the model is by keeping the positive news portion with maximum 60% and reducing contact with terrorist member.