Building a Regression Model to Predict Real Estate Sales Price- Linear Regression

January 3, 2017
Predict the real estate sales price of a house based upon various features about the house using Azure Machine Learning.
**Objective:** Predict sales price of a house based upon its various features. **Data:** Ames housing dataset includes 81 features and 1460 observations. **Feature Selection:** Features selections are most Important task we did here. Most relevant features are taken according to respons(Sales price). Here we use Pearson Correlation for test. **Categorical Casting:** Nominal categorical features were identified and cast to categorical data types using the meta data editor to ensure proper mathematical treatment by the machine learning algorithm. **Cleaning Missing Data:** Missing value were removed by (removed entire row)this. **Algorithm:** Linear regression model. **Evaluation:** Cross validation will be used to evaluate the predictive performance of the model. **Parameter Tuning:** This model is built to get better mean Coefficient of Determination & minimize mean RMSE. **Further Improvement:** Try to minimize mean RMSE with some other features and to get better Coefficient of Determination.