Estimation of real estate appraisal values using artificial intelligence methods

The accurate and efficient valuation of real estate has become increasingly important with the acceleration of economic and technological developments. Traditional valuation methods, which often rely on subjective expert opinions and limited data, face various challenges in ensuring accuracy and consistency. This study aims to develop a model that can estimate the market values of real estate properties in a selected province in Anatolia with high precision. In this context, a novel approach that surpasses traditional methods is presented through the utilization of machine learning algorithms such as Decision Trees, Extreme Gradient Boosting Regression, Gradient Boosting Regression, K-Nearest Neighbor Regression, Linear Regression, and Random Forest Regression. During the data collection process, a comprehensive dataset encompassing various real estate attributes was prepared and subsequently processed to make it suitable for statistical modeling. The model development and evaluation stages employed the Python programming language and its relevant libraries. The performance of each algorithm was tested using appropriate statistical metrics, and the most effective prediction method was determined. The findings reveal that ensemble methods, particularly Random Forest and Gradient Boosting Regression, outperform traditional valuation techniques in terms of accuracy and reliability. Artificial intelligence-assisted models demonstrate the potential to provide more informed decision-making processes through rapid and data-driven analyses independent of human intervention. Based on the study's results, it is recommended that AI-based methods be more widely adopted in real estate valuation practices in Turkey to enhance valuation accuracy and contribute to market transparency. Policymakers are advised to support regulations that integrate AI into valuation standards. Future research should focus on utilizing broader datasets, including real-time market data, and experimenting with deep learning models for more robust predictions. Additionally, ethical considerations such as data privacy and algorithmic fairness should be prioritized in these regulations.