Team Envible Analyze This 2017

Repository for American Express Analyze This 2017 challenge

Competition

The aim of the competition was to offer the students a pedagogical and professional experience in the analytics industry and provide an opportunity to explore analytics from a practical perspective.

Data

The dataset contained user data and we had to predict

1. If the user will buy a credit card
2. And which credit card will he buy

Approach

• Our approach was to first perform binary (buying a card or not) classification and then multiclass (buy which card) classification.
• The reasoning for the approach taken was that the data was highly skewed. The ratio of users not buying a card to buying category 1 card to category 2 card to category 3 card was 30 : 3.5 : 2.8 : 2.7.
• This two step approach helped us takle the problem of skewed data to some extent.

Problems faced

• Highly skewed data
• We were not able to perform predictions on the XGboost model due to a bug in the H2O framework and time constraint
• Limited time to use all the data analysis findings in the machine learning model building

Result

• Our final logloss on cross-validation using GBM was around 0.73
• Using XGboost for multiclass classification logloss was around 0.71
• The final submission made use of GBM for both binary(card or no card) and multiclass(which card) classification