# ldpseg

**Repository Path**: yang-joseph/ldpseg

## Basic Information

- **Project Name**: ldpseg
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 5
- **Created**: 2025-03-27
- **Last Updated**: 2025-03-27

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# AMOUE

*AMOUE: Adaptive modified optimized unary encoding method for local differential privacy data preservation*

This repository is the official implementation of the paper:

[AMOUE: Adaptive modified optimized unary encoding method for local differential privacy data preservation - ScienceDirect](https://www.sciencedirect.com/science/article/abs/pii/S0045790624007183)

## Citations

The details of this pipeline are described in the following paper. If you use this code in your work, please kindly cite it. Thanks.

```bash
@article{gao2024amoue,
  title={AMOUE: Adaptive modified optimized unary encoding method for local differential privacy data preservation},
  author={Gao, Tianchong and Fu, Hailong and Wang, Shunwei and Zhang, Niu},
  journal={Computers and Electrical Engineering},
  volume={120},
  number={Part B},
  pages={109791},
  year={2024},
  publisher={Elsevier},
  doi={10.1016/j.compeleceng.2024.109791}
}
```

## Installation

```bash
pip install -r requirements.txt
```

## Result

The following are our experimental results. We tested our methods under the conditions of different datasets and different privacy budgets.

![result](resources/table.png)

## Instructions

```bash
python MNIST.py --perturbation_method=AMOUE --save_data=True
python MNIST.py --perturbation_method=MOUE --save_data=True
python MNIST.py --perturbation_method=OUE --save_data=True
```