Predictive Analytics of Housing Markets Using Random Forests and Multidimensional Urban Indicators
Abstract
In the past few years, the housing market in rapidly urbanizing regions such as Uttarakhand has seen significant change due to increased infrastructure spending and shifts in urban development policies. Accurate forecasting of house prices matters not only for future homebuyers and investors but also for city planners and government authorities attempting to provide sustainable development. This research study sets forth an integrated machine learning-based model based on Random Forest Regression in order to predict and model the prices of housing in four key cities of Uttarakhand, namely Dehradun, Haridwar, Haldwani, and Rudrapur. The study begins with the collection of relevant datasets that include variables such as historical housing prices, investment in infrastructure, and urban development indicators. Next is data preprocessing and feature selection, followed by training and validation of Random Forest models using MATLAB. The simulation results indicate that the models perform optimally in all the cities, with the lowest Root Mean Squared Error (RMSE) in Dehradun (10.869), implying a higher degree of uniformity in the housing market of the capital city. Visual inspection of trends in urbanization development and infrastructure investment also validates the model prediction results. The study demonstrates the capability of Random Forest Regression to capture complex, nonlinear patterns in real estate data. The study presents a reliable and scalable model for urban housing price estimation, which provides an entry point to more strategic and knowledgeable urban development programs