Penerapan Metode Self-Organizing Map (SOM) Untuk Visualisasi Data Geospasial Pada Informasi Sebaran Data Pemilih Tetap (DPT)

Abstract

A Self-Organizing Map (SOM) is a neural network method that has been introduced by Professor Teuvo Kohonen since 1980, as an artificial neural network topology without supervision (Unsupervised ANN), in which the learning process does not require supervision or target output. In this paper, the implementation methods of SOM is used to perform data clustering voters (DPT) in the General Election Governor level. The results of the clustering process then used to perform data classification air-geographical references (geo-referenced) that integrates visualization of the output space through a cartographic representation of the color settings, and explore the use of line width between the boundary-water element geographic reference, calculated according to the distance in the input space is best between locations. Clustering results is then used as the basis for determining the color criteria in the development of the Web-GIS application based on interval number of voters.