Implementation Of The K-Means Cluster Algorithm In Rice Production Mapping And As A Decision Support For Agricultural Function Transition

Abstract

According to the Food and Agriculture Organization of the United Nations (FAO) in 2014 Indonesia ranked 3rd with a total rice production of 70.6 million tons, but it still remains a rice importing country. As one of the districts known as rice barn, Sukoharjo is targeted to continue to increase crop productivity every year to keep up with the growing population, so it is necessary to know areas with less optimal yields, and minimize changes in agricultural land use change. A mapping method for harvest results is needed to group data in each region based on the similarity of harvest data. In data mining, clustering techniques are known that can be used to map harvest productivity data based on their similarity. This study applies clustering techniques using the KMeans algorithm to map rice harvest productivity data by dividing data into 3 groups, namely many, medium, and less. The research method used is SDLC (Software Development Life Cycle) with a waterfall model. The K-Means algorithm is implemented using website-based programming to map harvest productivity data using attributes of planting area and rice production. The results of the mapping are visualized into a recommendation of agricultural land clustering and agricultural products as well as one of the decision makers in the transfer of agricultural functions so that subdistricts that have a lot of productivity, are moderate and lacking based on the characteristics of the data