Skip to content

Implement a Homomorphic Encryption Logistic Regression model using SEAL library

Notifications You must be signed in to change notification settings

huyvu1404/he_logistic_regression

Repository files navigation

Logistic Regression Over Encrypted Dataset

Introduction

In this project, I implemented a logistic regression model capable of operating on an encrypted dataset using homomorphic encryption. This approach ensures data security while performing computations on sensitive data, addressing significant privacy concerns. The project involves designing algorithms, preprocessing data, and evaluating model performance while keeping the data encrypted throughout the process.

Main Tasks

  • Design Algorithms: Developed and designed the logistic regression algorithm tailored to work with encrypted data.
  • Data Exploration and Preprocessing: Explored and preprocessed dataset to prepare it for encryption.
  • Training and Evaluation: Train logistic regression models and evaluate their performance to compare average accuracy and AUC score.

Languages

  • Python

Tools

  • Jupyter Notebook

Packages

  • Microsoft SEAL
  • NumPy & Pandas

About

Implement a Homomorphic Encryption Logistic Regression model using SEAL library

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published