Average income predication
This Project demonstrates solving Income Predictions for datasets available in "https://www.kaggle.com/c/tcdml1920-income-ind/data". Various methodologies are followed in this project to clean the data, process the data and fit the data to the best available model to get the error rate of 57k. The problem statement for this project is available at "https://www.kaggle.com/c/tcdml1920-income-ind/overview".
Methodologies used include feature selection using EDA analysis.One-hot ecnoding to encode the categorical values.Tried Naive Bayes,CATBOOST models and found Naive bayes works well when the output label values are also transferred to their log values.
Main packages used where numpy,pandas and model libraries.RMSE is used as evaluation model.
Run model.py to run the program.