Zoran Gacovski , Josip Kolic , Rosica Dukova and Marko Markovski


In this paper we present an application of data mining techniques in order to make price prediction of the real estate properties in the city of Skopje. The current research on the real estate data is insufficient, resulting in misunderstandings between the key players - local government, construction companies, real estate agencies and the potential clients. A dataset from over 1000 transactions in the past three years was used. Five variables (attributes) for each apartment were taken into consideration. We have used SQL Server database and Microsoft Business Intelligence tool (three different algorithms – decision trees, neural networks and logistic regression) in order to perform the price prediction. Also, it can be useful for the prediction of future trends in the urban development of city of Skopje.


Data mining techniques, Real estate property value, Microsoft Business Intelligence.