Data Science Games 2018

This repository contains data exploration, preprocessing methods and model used for used to predict the probability that a given customer is going to express an interest in buying or selling a bond. Details of challenge: https://www.kaggle.com/c/DSG2018-qualifiers

Overview of Implementation

• Combine data across fields: user info, market, trade activity, bonds info
• Summarize market data into weekly trends
• Basic Preprocessing - Onehotencoding, Normalization, Lagging (used previous week activity to predict this weeks action)
• Model - Gradient Boosting Tree

Built With

• Python
• Sklearn
• Xgboost

Authors

• Jin Wei Tan

License

This project is licensed under the MIT License