Pemetaan Kerentanan Tingkat Kriminalitas Menggunakan Metode Self Organizing Map

Abstract

The crime rate in an area can be a serious problem and can endanger the safety and welfare of the community. One important aspect in handling crime problems is understanding and disclosing patterns and characteristics of criminal incidents. This understanding is the basis for formulating effective crime prevention and control policies. Methods that have been widely used include Self-Organizing Map (SOM). Self-Organizing Mapping (SOM) was introduced in 1982 by Teuvo Kohonen as an unsupervised learning method. SOM can perform fairly valid and unbiased mapping, allowing it to map feature-based datasets through self-organization rules. The SOM architecture or what is often called a coherent network is a network consisting of two layers, namely the input layer and the output layer. Every neuron in the input layer is connected to every neuron in the output layer. Each neuron in the output layer interprets the class of the input provided. Based on the research results, it was found that it was certain that the best cluster was in cluster 2 for districts/cities in the Special Region of Yogyakarta, Central Java and East Java. Cluster 1 consists of 15 districts/cities, and cluster 2 consists of 63 districts/cities