Skip to content

Latest commit

 

History

History
31 lines (17 loc) · 664 Bytes

README.md

File metadata and controls

31 lines (17 loc) · 664 Bytes

Classification Project

In this repository a classification project is presented. It is based on the Santander Customer Transaction Prediction Kaggle competition.

Binary Classification Problem:

Which customer will make a specific transaction in the future, irrespective of the amount of money transacted.

Dataset

You can download the dataset here: https://www.kaggle.com/c/santander-customer-transaction-prediction/data

Evaluated Metrics:

Precision Recall F1 Score Receiver Operating Characteristic Curve (ROC)

Technique for imbalanced data

Synthetic Minority Oversampling Technique (SMOTE)

Models

Logistic Regression Light GBM