Machine Learning, Deep Learning, and Hybrid Approaches in Real Estate Price Prediction: A Comprehensive Systematic Literature Review

The real estate refers to an extensive field that deals with the purchase, selling, or management of properties, and it stands out as an influential industry in the economic development process, indicating that the precise determination of the price is one of the most effective tools for decision-making among different subjects of the market and authorities. Price prediction improves investment plans, risk management, fair price transactions, and provides key inputs to economic and urban planning. This systematic review categorizes the existing approaches into three groups: machine learning, deep learning, and hybrid models, based on the selected literature from a broad search of numerous databases and using rigorous criteria. The review indicates that the traditional and current machine learning models have relatively high levels of predictive accuracy for small datasets. However, deep learning techniques are preferable for handling large and complex data, while hybrid models have even more potential to increase prediction accuracy. The present study indicate that these sophisticated techniques can enhance and enrich price forecasting models, which can be insightful to various industrial decision-makers and informative for future research endeavours.