Evaluating the Effectiveness of Random Forest versus Decision Tree Algorithms in Predictive Analytics for Enhancing Sustainable Development Strategies in Real Estate, Infrastructure, and Construction: A Machine Learning Perspective

This study explores the application of machine learning algorithms, specifically decision tree and random forest models, to predict house prices within the Indian real estate market. Utilizing a comprehensive dataset obtained from Kaggle, this research encompasses extensive data preprocessing, feature engineering, and exploratory data analysis to identify key predictors of house prices. The comparative analysis aims to assess the accuracy, efficiency, and applicability of both models in the context of predictive analytics for sustainable development goals (SDGs) in real estate. The findings reveal that the random forest model outperforms the decision tree model, demonstrating superior predictive accuracy with a lower Root Mean Squared Error (RMSE). Key predictors identified include living area, number of bathrooms, and the condition of the property, with the random forest model providing deeper insights into feature interactions and their collective impact on house prices. This research highlights the potential of advanced machine learning techniques in enhancing real estate market analysis and strategic decision-making. The superior performance of the random forest model underscores its value in predictive analytics, offering stakeholders in the real estate sector a robust tool for achieving greater accuracy in property valuation and market analysis. Future research directions include the integration of additional data sources and the exploration of other machine learning algorithms to further refine predictive models and extend their applicability to sustainable development within the real estate, infrastructure, and construction industries.