Icon

02_​Autoencoder_​for_​Fraud_​Detection_​Deployment

Keras Autoencoder for Fraud Detection - Deployment

This workflow applies a trained autoencoder model to detect fraudulent transactions.

Reading data, parameters, and network
The Fraud Detector

  • data normalization

  • reproduction of data by autoencoder

  • MSE calculation

  • Applying threshold

Taking Actions

IF trx is legitimate => do nothing
IF trx is fraud candidate => send email to owner

send email to owner
Send Email
Read Keras model creditcard_autoencoder.h5
Keras Network Reader
Define outliers by the best threshold for reconstruction error
Rule Engine
Math Formula
Table Row to Variable
Reconstruct deployment data with autoencoder
Keras Network Executor
threshold.table
Table Reader
Normalizer (Apply)
creditcard_deployment.csvRead one row at a time
CSV Reader
normalizer.model
Model Reader
Table Row to Variable
CASE Switch Start

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: Maarit Laukkanen
Created at: 2019-02-27
On NodePit since: 2020-12-17
Last update: 2026-11-10
Created with KNIME version: v5.4.2
Tags: autoencoderKerasneural networkbackpropagationdeep learningencoderdecoderreconstructioncredit card datafraud detectionanomaly detectiongenerative modeldeploymentpredictionfraudbankingcredit cardcybersecurity

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!