Need a Home? Start the Data Mining! A Data Mining Application in Weka on Real Estate Market

Using real data regarding the very active real estate market from Cluj-Napoca, Romania (a 500,000 people city), this project wants to help people in their decision of buying an apartment or evaluating it, finding a relation between it's characteristics and market value.The comparison of many apartments with many specifications and the right evaluation of their prices might be difficult. We used Weka, software based on Java, that uses algorithms specialized in extracting new knowledge and relations, not visible in a database. The dataset was extracted from a large database, provided by a newspaper. We applied a linear regression algorithm for evaluating the price and the J48 algorithm to make a decision tree.The algorithms found simple and efficient solutions in evaluating apartments, with a very low error. Therefore, Weka is a strong tool in analyzing the real estate market, can be used in the automation of evaluations and the extension of the analysis on larger markets