Predicting Duration of Residential Units’ Sale Using Machine Learning Techniques

Abstract

Real estate is vital to meeting basic needs and offering solid investments. Informed investment decisions require understanding home sale factors. Estimating residential property sales time can boost rewards and decrease risks. Two steps are involved in this paradigm. Significant characteristics are identified using filter weighting and regression techniques. K-nearest Neighbour, Naïve Bayes, and Decision Trees use characteristics identified in the first stage. Determining the most efficient model requires comparing their precision. The study provides brokers with insights for improved sales forecasting and transaction management. This technique allows stakeholders to adjust their plans based on market fluctuations, potentially leading to more profitable real estate investments. This research clarifies real estate investment strategies for greater returns.