AUTHOR=Elnaeem Balila Abdulsalam , Shabri Ani Bin 

TITLE=Comparative analysis of machine learning algorithms for predicting Dubai property prices

JOURNAL=Frontiers in Applied Mathematics and Statistics

VOLUME=Volume 10 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/applied-mathematics-and-statistics/articles/10.3389/fams.2024.1327376

DOI=10.3389/fams.2024.1327376

ISSN=2297-4687

ABSTRACT=Predicting property prices is a crucial task in the real estate market, and machine learning algorithms offer valuable tools for accurate predictions. In this study, we perform a comprehensive comparison of seven well-known machine learning algorithms, EEMD-SD-SVM, SVM, Gradient Boosting, Random Forest, KNN, Linear Regression, ANN, and Decision Trees, in the context of predicting property prices in Dubai. Our evaluation is based on four key performance metrics:  R2, MSE, RMSE, and MAPE. The primary objective of this comparison is to assess the predictive performance of these algorithms specifically within the Dubai property market. MSE and RMSE provide insights into the prediction errors, MAPE quantifies the accuracy of predictions in percentage terms, while R2 is a statistical measure that represents the proportion of the variance in the dependent variable (property prices) that can be explained by the independent variables (features) in the predictive model. Our findings shed light on the strengths and limitations of each algorithm for this specific task. EEMD-SD-SVM and SVM may excel in scenarios where precise decision boundaries are vital, while Gradient Boosting and Random Forests demonstrate robust performance when handling complex and noisy property price data. KNN proves valuable in capturing localized patterns, Linear Regression for straightforward regression tasks, ANN for deep learning with extensive datasets, and Decision Trees for models that offer interpretability in understanding the factors influencing property prices. The study underscores the significance of model tuning, feature selection, and data preprocessing to enhance the predictive power of these algorithms. Additionally, it discusses the practical aspects of computational efficiency, model interpretability, and their suitability for scalability in real-world applications. Ultimately, this comparative analysis provides valuable guidance for real estate professionals, data scientists, and stakeholders interested in selecting the most suitable machine learning algorithm for predicting property prices in Dubai, with a focus on the essential evaluation metrics of MSE, RMSE, MAPE, and R2. The paper provides valuable insights into the applicability and performance of different machine learning algorithms for predicting property prices in the specific context of Dubai, aiding stakeholders like real estate agents, buyers, sellers, or investors in making informed decisions.