Icon

Projeto

Keras Autoencoder for Fraud Detection - Training

This workflow trains an autoendcoder model to detect fraudulent transactions.

Keras Autoencoder Architecture
Data Preprocessing
Training the Autoencoder
Optimizing threshold K
Final Performance
Data Cleaning and Preparation

Keras Autoencoder Architecture
Data Preprocessing
Training the Autoencoder
Optimizing threshold K
Final Performance
Data Cleaning and Preparation

Counter Generation
GroupBy
Joiner
Math Formula
2/3 of negatives for training
Table Partitioner
Bar Chart
Rule Engine
Math Formula
1/3 of negativesand 1/3 positivesfor validation
Concatenate
Pie Chart
Apply network
Keras Network Executor
Rule Engine
Min-max normalization
Normalizer
Units: 16Activation: Sigmoid
Keras Dense Layer
1/3 of negativesfor validation
Table Partitioner
Rule Engine
Normalizer (Apply)
Units: 29Activation: Sigmoid
Keras Dense Layer
Date&Time Part Extractor
Units: 26Activation: Sigmoid
Keras Dense Layer
One to Many
Apply network
Keras Network Executor
Read credit card data
CSV Reader
Classifytransactions based onthreshold
Rule Engine
Class
Number to String
Math Formula
Top:Class = 0
Row Splitter
Units: 8Activation: Sigmoid
Keras Dense Layer
Normalizerfor deployment
Model Writer
Rule Engine
Write model
Keras Network Writer
Rule Engine
10 % for validation
Table Partitioner
Bar Chart
Pie Chart
Units: 23Activation: Sigmoid
Keras Dense Layer
Train with Loss function=MSE Optimizer=Adam
Keras Network Learner
2/3 of negatives for training
Table Partitioner
1/3 of negatives and all positives for validation
Concatenate
Min-max normalization
Normalizer
Variable to Table Row
Table Writer
Normalizer (Apply)
Shape: 23
Keras Input Layer
Missing Value
Units: 29Activation: Sigmoid
Keras Dense Layer
Date&Time Part Extractor
Units: 8Activation: Sigmoid
Keras Dense Layer
String to Date&Time
Units: 8Activation: Sigmoid
Keras Dense Layer
Rule Engine
Variable to Table Row
Units: 16Activation: Sigmoid
Keras Dense Layer
Rule Engine
Table Writer
10 % for validation
Table Partitioner
Units: 23Activation: Sigmoid
Keras Dense Layer
Units: 8Activation: Sigmoid
Keras Dense Layer
Units: 26Activation: Sigmoid
Keras Dense Layer
One to Many
String to Date&Time
Train with Loss function=MSE Optimizer=Adam
Keras Network Learner
Shape: 23
Keras Input Layer
Write model
Keras Network Writer
Missing Value
Classifytransactions based onthreshold
Rule Engine
Scorer (JavaScript)
Scorer (JavaScript)
Top:Class = 0
Row Splitter
String to Number
Counter Generation
Column Renamer
GroupBy
Column Filter
Rule Engine
Joiner
String Manipulation
Read credit card data
CSV Reader
Math Formula
Rule Engine
Normalizer (Apply)
String to Number
String Manipulation
Threshold Optimization
Rule Engine
Normalizer (Apply)
Normalizerfor deployment
Model Writer
Threshold Optimization
Class
Number to String
Column Filter
Column Renamer

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: 2019-02-25
On NodePit since: 2026-05-12
Last update: 2026-05-12
Created with KNIME version: v5.11.0
Tags: autoencoderKerasneural networkdeep learningencoderdecoderreconstructioncredit card datafraud detectionanomaly detectionmodel evaluationbankingcredit cardfraudcybersecurity

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

Try NodePit Runner!