• Stars
    star
    214
  • Rank 184,678 (Top 4 %)
  • Language
    Jupyter Notebook
  • Created over 6 years ago
  • Updated about 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

2nd Place Solution 💰🥈

Home-Credit-Default-Risk

LeaderBoard: https://www.kaggle.com/c/home-credit-default-risk/leaderboard

Discussion: https://www.kaggle.com/c/home-credit-default-risk/discussion/64722

I was in charge of feature engineering in this team. Private LB Model Pipeline

Following are the features I made

Features

1. Current Application

  • Simple calculation(delta, ratio, etc...)

2. Previous Application

  • Aggregated features(all, 1year, 2years, 3years)
  • last and first application

3. POS CASH balance

  • Aggregated features(all, 1year, 2years, 3years)

4. Installments Payments

  • Aggregated features(all, 1year, 2years, 3years)

5. Credit Card Balance

  • Aggregated features(all, 1year, 2years, 3years)

6. Bureau

  • Aggregated features(all, 1year, 2years, 3years)
  • last and first application

7. Bureau Balance

  • Aggregated features(all, 1year, 2years, 3years)

8. Other

  • Aggregated features using above data(all, 1year, 2years, 3years)