Potential classification of Smart Village – Smart Economy with Deep Learning methods

Abstract

Development gap in the city and village is still happening on Indonesia. It happened because of the massive urbanization factors. Poverty in the Indonesian villages are relatively higher than on the urbans. In order to reach the maximal city development, Ministry of Village, Development of Disadvantaged Regions and Transmigration of Indonesia created a sustainable village development program namely Village’s Sustainable Development Goals (SDGs) and optimized the village potential data. This study aimed to design the smart village – smart economy classification system by using deep learning methods on village potential data on Indonesia at 2020. The method used in this study is data mining processes namely KDD (Knowledge Discovery and Data mining). The result in this study showed the best models were obtained which consisting of 2 hidden layers and each layer is 128, 128 layers which using target class from the process of calculating the score is able to reach 94.93% of the accuracy from the training process and 96% on the testing process and succeeded to classify the potentials of smart village – smart economy.