IMPLEMENTATION OF ARTIFICIAL NEURAL NETWORK FOR STATE – OWNED ASSETS FORECASTING OF ROOM RENTAL PRICES IN INDONESIA

Abstract

Leasing is a state-owned assets utilization scheme that needs to be optimize because of its easy to find objects and large potential for non-tax revenue. In the city of Yogyakarta, the economy grows above the national average, this is supported by the mobility of tourists, overseas students, and businessman. The characteristics of the regional economy are suitable for the optimization of state-owned assets through leasing scheme in the form of lodging room. The author tries to develop a state-owned assets leasing price forecasting model for lodging room using an Artificial Neural Network to capture the potential state revenue. By using market data for lodging room rental from the OYO website, author create a model architecture with the backpropagation algorithm. Analysis results of this study indicate that the obtained network model achieves an accuracy of 97.5%. There are 25 state-owned assets buildings that can be projected as objects of lodging space rental utilization with a predicted rental value of IDR 108,570.00 to IDR 122,669.00 per day.