# Data-poisoning-attacks-on-factorization-based-collaborative-filtering

**Repository Path**: xtzgg/Data-poisoning-attacks-on-factorization-based-collaborative-filtering

## Basic Information

- **Project Name**: Data-poisoning-attacks-on-factorization-based-collaborative-filtering
- **Description**: https://github.com/fuying-wang/Data-poisoning-attacks-on-factorization-based-collaborative-filtering.git
- **Primary Language**: Python
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2022-04-11
- **Last Updated**: 2022-04-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Data-poisoning-attacks-on-factorization-based-collaborative-filtering
# Reference
* "Data poisoning Attack on Factorization-Based Collaborative Filtering" NIPS 2016 
* Github respository : "recommend"
* Github respository : "movielens"
# Illustration
* ALS_optimize.py: compute optimize solution using malicious data and normal data.
* ALS_optimize_old.py: compute optimize solution only using normal data.
* compute_grad.py: compute two parts of gradient using the formulations in the paper.
* dataset.py: functions of load dataset and build user item matrix.
* evaluation.py: predict data and compure RMSE
* main.py: the implement of PGA

wget https://files.grouplens.org/datasets/movielens/ml-20m.zip

修改main.py line20 file path

修改dataset.py 读取csv文件获取评分数据的代码