House pricing with data mining techniques

In this study a database from the real estate website has been used. The aim is to detect several important factors which will directly affect the price of housing. The organization of this work follows. First the database will be introduced, then the analyse results which are constructed by using KNIME and WEKA will be discussed and finally the prices predicted by the decision-trees will be analyzed. This work predicts price range of a housing by using the following steps. First various analyses are performed by using decision trees. Then clustering is performed by using K-means and Fuzzy C-means algorithms. Afterwards several important factors which affect housing prices are predicted and by using the results, the application is improved. Thus the application is now able to define the price range of any house.