2017 Seed Cup Semi-Final Competition
-
Firstly, we need to extract inherent features and of users and products. Furtherly, we extracted behavioural features of users and the interactive features of products all from user's behaviors
-
We developed an User-Product Purchasing Model
- We adopt user-based and product-based collaborative filtering to generate recommend products for users
-
macOS 10.14 High Sierra
-
Python 2.7.10
-
XGBoost
-
Pytorch
-
Scikit-Learn
The code commited on this repository might not work much well as expected, because we adopted many intermediate steps and operations to collect the best result!
This project is under the MIT Licence