The capstone aims to provide real estate investors suitable portfolios for investment in Dubai, considering maximum return on investment as per their investment goals. CRISP-DM methodology was used as it is a structured approach for planning a data mining project. The datasets from Dubai Land Department, RERA and Property Finder was used to conduct Exploratory Analysis, Predictive Analysis using Linear Regression, Random Forest Regression, Decision Tree Regression, Gradient Boost Regression and Time Series analysis using Exponential Smoothing, Holt forecasting, Holt Winter Method, ARIMA and SARIMA. To facilitate investors in easily searching for suitable portfolio to invest in based on their preference, a web platform has been designed. This platform also provides the option of using Predictive Analysis for estimating the returns. The best result for Predictive Analysis was provided by Gradient Boost Regression and for Time Series Analysis was provided by SARIMA model as the dataset includes seasonality.
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research (Dubai)
Krishna, Nikhil, "Dubai Real Estate Investment: A Predictive and Time Series Analysis" (2021). Thesis. Rochester Institute of Technology. Accessed from