GEOGRAPHIC INFORMATION SYSTEM OF EARTHQUAKE RISK ASSESSMENT USING K-MEANS ALGORITHM AND MEVN STACK CASE STUDY OF INDONESIA REGION

Abstract

Indonesia is located in a very active seismic region, known as the ring of fire. Earthquake risk assessment serves to categorize the level of hazard from earthquakes based on their effects. Earthquake risk assessment can be done using the clustering method. Clustering is the process of grouping a pattern that does not yet have a label and is done without supervision into a group that has certain characteristics. The algorithm used for clustering in this research is the K-Means algorithm. This algorithm allows data to be grouped into clusters so that data that have similarities are in the same cluster. The K-Means clustering model was developed using the magnitude and depth attributes of the earthquake. The model was successfully implemented and produced 15 clusters as the best number of clusters. The clustering results obtained from the model were then implemented in the Geographic Information System using MEVN Stack. MEVN Stack is a combination of framework and database consisting of MongoDB, Express.js, Vue.js and Node.js. This system can identify earthquake-prone areas by presenting information in the form of heatmaps displayed on a map. Each heatmap can show statistics complemented by historical information on earthquakes that have affected the area. In addition, the system allows users to measure earthquake risk through the Risk Map feature, where users can select a point on the map to see the level of earthquake risk at the point selected by the user.