Open AccessExploratory Data Analysis Towards Terrorist Activity in Indonesia Using Machine Learning Techniques
Author(s) -
Green Arther Sandag
Publication year - 2019
Publication title -
abstract proceedings international scholars conference
Language(s) - English
Resource type - Journals
ISSN - 2476-9606
DOI - 10.35974/isc.v7i1.1628
Subject(s) - terrorism , action (physics) , national security , exploratory research , feature selection , political science , computer security , criminology , psychology , computer science , sociology , law , artificial intelligence , social science , physics , quantum mechanics
Terrorism Activity is the subject of the talks in various countries, especially in Indonesia. Theactivities of terrorism are carried out in various ways using suicide bombs, violent action thataimed to demoralize by creating fear to the society and national security. In Indonesia,according to Kompas news website recorded there were 10 suicide bombings occurred in thepast 6 years and took many casualties in every event. With this, it certainly gives a threat tothe people in Indonesia in terms of physical, moral and even in terms of national security. Toovercome this problem, it is necessary to increase the national security so that terrorism canbe prevented, and it will not happen again. This study is aimed to conduct an exploratory dataanalysis and predict terrorist activity in Indonesia using K-Nearest Neighbor (KNN), and ¬kfold cross-validation. In this research, data selection, data cleaning, data reduction wascarried out and feature selection process which aimed to find out the most influential dataattributes. Based on the result of the analysis to predict the terrorist activity, the result of theaccuracy was obtained with a value of k = 8 at 88.86%.