Skip to content

to study bias and fairness in recommender systems have focused on improving fairness and mitigating bias only in situations and for items where a history of the user profile already exists. In this project, we explore the bias against new items without any feedback history which are added to recommender systems.

Notifications You must be signed in to change notification settings

nikhilvenkatkumsetty/Mitigating-Unfairness-and-Bias-in-Cold-Start-Recommenders

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Instructions to run the project

Create an anaconda environment with the command - conda create -n "env_name" python==3.7.5

activate the environment using the command - conda activate env_name

Install Tensorflow-1.14.0 package using the command - conda install -c conda-forge tensorflow=1.14.0

Install all the packages from 'requirements.txt' file using the command - pip install -r requirements.txt

Now, Navigate to Gen and Scale folder and run the command - python main.py to run the Joint-Learning generative model and Score-scaling model respectively.

Now, navigate to 'Data/ml1m' folder and run 'cold_bias_analysis_ColdRec.ipynb' and 'cold_bias_analysis_Debias.ipynb' files in Jupyter Notebook.

About

to study bias and fairness in recommender systems have focused on improving fairness and mitigating bias only in situations and for items where a history of the user profile already exists. In this project, we explore the bias against new items without any feedback history which are added to recommender systems.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published