Kaggle's "House Price Prediction" using Regression techniques.
This notebook is created to predict house prices based on Ames Housing dataset. The Kaggle competition is found here
File descriptions:
- train.csv - the training set
- test.csv - the test set
- data_description.txt - full description of each column, originally prepared by Dean De Cock but lightly edited to match the column names used here
For arriving to the predictions, I have used a weighted mean of the results of the following models:
- Lasso Regression
- Elastic Ridge Regression
- Kernel Ridge Regression
- Gradient Boosting Regression
- XGBoost
- LightGBM
The next step in improving the results would be adding the model stacking to the solution.