https://github.com/kennethleungty/anomaly-detection-pipeline-kedro

Anomaly Detection Pipeline with Isolation Forest model and Kedro framework
https://github.com/kennethleungty/anomaly-detection-pipeline-kedro

anomaly anomaly-detection credit-card credit-card-fraud data-science data-science-pipeline financial financial-data fraud fraud-detection kedro machine-learning machine-learning-pipeline ml mlops pipelines quantumblack

Last synced: 3 months ago
JSON representation

Anomaly Detection Pipeline with Isolation Forest model and Kedro framework

• Host: GitHub
• URL: https://github.com/kennethleungty/anomaly-detection-pipeline-kedro
• Owner: kennethleungty
• Created: 2022-03-11T15:37:27.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2022-12-27T02:43:58.000Z (about 2 years ago)
• Last Synced: 2024-10-29T05:34:50.288Z (4 months ago)
• Topics: anomaly, anomaly-detection, credit-card, credit-card-fraud, data-science, data-science-pipeline, financial, financial-data, fraud, fraud-detection, kedro, machine-learning, machine-learning-pipeline, ml, mlops, pipelines, quantumblack
• Language: Python
• Homepage: https://neptune.ai/blog/data-science-pipelines-with-kedro
• Size: 239 KB
• Stars: 23
• Watchers: 2
• Forks: 7
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Building and Managing an Isolation Forest Anomaly Detection Pipeline with Kedro

## Overview

Anomaly (fraud) detection pipeline on credit card transaction data using Isolation Forest machine learning model and Kedro framework

Link to article: https://neptune.ai/blog/data-science-pipelines-with-kedro

## Objective

Develop a data science pipeline to detect anomalous (fradulent) credit card transactions with the use of:

- **Isolation Forest** machine learning model - For unsupervised anomaly detection

- **Kedro** - An open-source Python framework for creating reproducible, maintainable, and modular data science code. This framework helps to accelerate data pipelining, enhance data science prototyping, and promote pipeline reproducibility.)

## Motivation

- Explore how unsupervised anomaly detection works, and better understand the concept and implementation of isolation forest

- Leverage Kedro framework to optimally structure data science pipeline projects

## Data

The [credit card transaction data](https://github.com/Fraud-Detection-Handbook/simulated-data-transformed) is obtained from the collaboration between Worldline and Machine Learning Group. It is a realistic simulation of real-world credit card transactions and has been designed to include complicated fraud detection issues.

## General Pipeline Structure

![Alt text](/docs/images/01_DS_Pipeline_Overview.png?raw=true)

## Anomaly Detection Pipeline Structure

![Alt text](/docs/images/05_Anomaly_Detection_Pipeline_Blueprint.png?raw=true)

## Steps

1. Change path to project directory in command line - `cd C:/Anomaly-Detection-Pipeline-Kedro`

2. Initialize Conda virtual environment (create one if not done so) - `conda activate env_kedro`

3. Execute a pipeline run with `kedro run`

Please see the [walkthrough article](https://neptune.ai/blog/data-science-pipelines-with-kedro) for details