Data Mining System to Analyze Hotspot Data

This paper was not presented at the conference.

Abstract

This paper was not presented at the conference.

Forest and land fires in Indonesia have increased in intensity and spread. One of the activities to prevent and reduce the occurrence of this phenomenon is to analyze hotspot data. The hotspot data will be analyzed using data mining techniques, namely spatio-temporal clustering and sequential pattern mining. Both systems are already available but are still used separately. This study aims to develop a system that combines the two systems into one unit to analyze the hotspot data more efficiently. The method used in this research is prototyping. The data for this system uses the API from SiPongi which has been modified. The system is a web application using the Shiny framework. This system makes it easy for users to perform cluster and sequence analysis of hotspots in the same system. Visualization of clusters and sequences is given in the form of maps with markers, tables, and parameters to get results according to user preferences. The test results use black box testing and resulting in 94% effectiveness.