Regression: Resale Price Prediction

July 10, 2015
Using historical transcated data, this experiment predicts house resale price using regression algorithm.
This is a resale price model which predicts resale price of houses . This model uses linear and boosted decision tree regression algorithm and evaluate results of chosen algorithms. About the Dataset Singapore Housing and Development Board releases data of Resale prices of flats, transacted from 2013 onwards. http://data.gov.sg/Metadata/SGMatadata.aspxid=2315050000000014695E&mid=171158&t=TEXTUAL "The datasets provided by the Singapore Government and its Statutory Boards via Data.gov.sg are governed by the Terms of Use available at www.data.gov.sg/common/terms.aspx. To the fullest extent permitted by law, the Singapore Government and its Statutory Boards are not liable for any damage or loss of any kind caused directly or indirectly by the use of the datasets or any derived analyses or applications" Insights It primarily shows usage of feature selection using Azure ML components Filter Based Feature Selection. Pearson Correlation, Chi Squared feature scoring methods were evaluated. For the dataset, Chi Squared was preferred with 4 independent variables. About the Author : Saurabh Singh is a Data Science Evangelist @ Cognizant Technology Solutions