Skip to content

pyy0715/Neural-Collaborative-Filtering

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Neural Collaborative Filtering

Pythorch Version of Neural Collaborative Filtering at WWW'17

model

Paper

Official_Code(Keras)

Author: Dr. Xiangnan He

Keypoints

  • The problem that the thesis intends to solve is to recommend the item to the user based on implicit feedback.
  • Applying deep learning to user-item interaction in matrix factorization
  • Using a network structure that takes advantage of both dot-product (GMF) and MLP
  • Use binary cross-entropy rather than MSE as loss function
  • Use point-wise loss + negative sampling

Results

Screenshot

Best epoch 005: HR = 0.677, NDCG = 0.399

Requirements

python==3.7.7 pandas==1.0.3 numpy==1.18.1 torch==1.4.0

Reference