Hybrid Variational Autoencoder-based Models for Fraud Detection

Hybrid Variational Autoencoder-based Models for Fraud Detection

Hybrid Variational Autoencoder-based Models for Fraud Detection. Typical anomaly involves highly imbalanced datasets

Introduction

The objective of this work is to develop deep learning models using Keras/Tensorflow API to detect anomalous credit card transactions and classify fraud. Typical anomaly detection involves highly imbalanced datasets. We employ stacked variational autoencoders (VAE) in an unsupervised setting to efficiently classify fraudulent transactions. The models are tested using Kaggle credit-card fraud dataset. Additional benchmarking is done using KDDCUP99–10% dataset as well.

Stacked variational autoencoders (VAEs) are used to learn latent space representation of “normal” credit card transactions by training them only with “normal” data. The anomalous transactions are identified by calculating the reconstruction error using the trained VAE network. Unusually high reconstruction errors are indicative of anomalous transactions/fraudulent transactions. We identify an optimum threshold for reconstruction error, beyond which the transactions are labelled as fraud, as the one which yields highest model accuracy (in terms of F1-score, and area under ROC).

We also explore different hybrid models involving combination of VAEs with supervised learning models such as Random-Forest Classifiers in order to improve the classification accuracy. In the hybrid workflow, the stacked VAEs are used solely as generative models to augment the under-sampled data (i.e. anomalies). The VAE-based data augmentation is used to boost the performance of a Random-Forest classifier.

deep-learning variational-autoencoder fraud-detection tensorflow kl-divergence

Bootstrap 5 Complete Course with Examples

Bootstrap 5 Tutorial - Bootstrap 5 Crash Course for Beginners

Nest.JS Tutorial for Beginners

Hello Vue 3: A First Look at Vue 3 and the Composition API

Building a simple Applications with Vue 3

Deno Crash Course: Explore Deno and Create a full REST API with Deno

How to Build a Real-time Chat App with Deno and WebSockets

Convert HTML to Markdown Online

HTML entity encoder decoder Online

Applying Anomaly Detection with Autoencoders to Fraud Detection

Applying Anomaly Detection: Credit card fraud can be classified as an anomaly and using autoencoders implemented in Keras it is possible to detect fraud.

Top Deep Learning Development Services | Hire Deep Learning Developer

Inexture's Deep learning Development Services helps companies to develop Data driven products and solutions. Hire our deep learning developers today to build application that learn and adapt with time.

Fraud detection — Unsupervised Anomaly Detection

An 100% unsupervised approach to discover frauds on credit card transactions. One of the greatest concerns of many business owners is how to protect their company from fraudulent activity.

Top 10 Deep Learning Sessions To Look Forward To At DVDC 2020

Looking to attend an AI event or two this year? Below ... Here are the top 22 machine learning conferences in 2020: ... Start Date: June 10th, 2020 ... Join more than 400 other data-heads in 2020 and propel your career forward. ... They feature 30+ data science sessions crafted to bring specialists in different ...

Credit Card Fraud Detection via Machine Learning: A Case Study

Credit Card Fraud Detection via Machine Learning: A Case Study. A machine learning guide on how to identify fraudulent credit card transactions by using the PyOD toolkit.