Design and Implementation of Residential Rental Rates Forecast Model using Data Mining Algorithms

This study examines the impact of the hood structures and characteristics on the residential rental pricing in Petaling Jaya, Selangor. The absence of analytical forecasting tools for this region is the main driver of this research. We propose a systematic pipelined approach composed of a geographic data handler, rental analytics unit, and a price-based predictive module. The latter offers a regional-aware real estate market forecast. The analytical part is equipped with chart indicators paired to illustrate recent pricing trends in the selected regions. The geographic data handler visualizes the local amenities on a map-based tool integrating listing services taken from iBilik, a well-known real estate website. The modules are acting a coherent framework assisting the public in understanding market trends and demands. The experimental result of comparing different machine learning algorithms shows that XGBoost has the most accurate performance prediction among the others.