Skip to content

mitao-cat/feature-level_bias

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

feature_level_bias

This is the implementation of our paper in WWW'24 (short):

Understanding and Counteracting Feature-Level Bias in Click-Through Rate Prediction

By Jinqiu Jin, Sihao Ding, Wenjie Wang, and Fuli Feng

You can also refer to the long paper version by this link.

  1. To reproduce results in normal biased tests:

    cd ./code
    python reduction.py --dataset ml-1m --model nfm --alpha 0
    # dataset: {'ml-1m','book','kuairand'}
    # model: {'fm', 'nfm'}
    # alpha: 1->basemodel; 0->reduction
  2. To reproduce results in debiased tests:

    cd ./code
    # model: {'fm', 'nfm'}
    # reconstruction: 0->basemodel; 1->reconstruction
    python reconstruction.py --reconstruction 1 --model nfm

Reference: shenweichen/DeepCTR-Torch - GitHub

Releases

No releases published

Packages

No packages published

Languages