Comparison Data Mining based on Optimization Algorithms in Receiving Electricity Subsidies

Abstract

Electricity subsidy is the Indonesian Government program through PT. PLN (Persero) to the underprivileged community. Hopefully this program can ease the burden of society due to the economic impacts that occur. There is a problem in the distribution process that has been happening is not on target, it certainly resulted in a financial loss that has been issued by the Government and people who do not feel government assistance. This research uses a comparison of Data Mining classification method based on optimization algorithm to get the best model. Solutions that are expected to assist the Government in reducing financial losses in the delivery process are occurring as well as providing benefits for the government to properly process the distribution of targets. Research provides results that the C4.5 method with the Genetic Algorithm (GA) optimization is a better level of accuracy than Particle Swarm Optimization (PSO) and Differential Evolutin (DE) with a value of 92.28% accuracy.