Ruth Ema Febrita, Adyan Nur Alfiyatin, Hilman Taufiq, Wayan Firdaus Mahmudy
House prices in Indonesia tend to increase every time. This is due to the increasing demand for residential sector every year, especially in urban areas. Prediction of house prices is important, especially for property investors and potential buyers, so they can make careful planning related to home sales and purchases. House price prediction also useful for banker in assessing asset prices as a references in loan approval. There are several factors affect the house price, including physical attributes, concepts, locations, as well as several economic factors prevailing at that time. Fuzzy inference system can be used to make predictions. However, the difficulty of applying the fuzzy inference system is establishing the membership function and choosing the rules to be used. This research goal is to extract fuzzy rules (membership functions and inference rules), which can be used to predict house prices based on nearby objects location. K-Means clustering method is used to extract inisial values to form fuzzy membership functions and inference rules of several groups of residential. This research produces a good-interpretability fuzzy system shows a satisfactory result of predictions. © 2017 IEEE.
Brawijaya University, Faculty of Computer Science Malang, Indonesia
