Assessing Machine Learning Models for Predictive Analytics in Urban Rental Markets: Rent Price Forecasting in India

This research marks a groundbreaking application of a diverse array of machine learning techniques to predict rental house prices specifically in the major Indian cities of Pune, Delhi, and Mumbai. These urban areas are known for their rapid economic growth and complex demographic shifts, which introduce significant variability into the rental market. Moving beyond the conventional focus on property sales, this study explores the rental market using advanced machine learning models such as Linear Regression, Ridge, Lasso, ElasticNet, Random Forest, Gradient Boosting, K-Neighbors, Decision Tree, Support Vector Machines, and Polynomial Regression. This novel approach is tailored to meet the unique challenges of the Indian rental landscape, offering a detailed examination that transcends traditional analytic methods. By meticulously adapting these models to the nuanced dynamics of rental prices in Pune, Delhi, and Mumbai, the study fills a crucial gap in existing real estate research. Our results highlight the varied performance of these models, with non-linear methods like K-Neighbors and Random Forest emerging as particularly effective in capturing the complex interplays affecting rental prices. This research not only deepens the understanding of machine learning's utility in real estate analytics but also provides vital data-driven insights to stakeholders involved in policymaking, investing, and managing within India's vibrant and evolving rental markets.