CSV Reader - original dataset - 284,807 transactions
Missing Value - removes missing value to clean data
Number to String - converts class column into categorical data for use in the regression learner
X-Partitioner - splits data for sampling and model training
Row Filter - sorting for just the fraudulent transaction through the binary label (1)
Independent Sample T-Test - whether the mean transaction Amount differs between fraud and legitimate transactions (null hypothesis rejected)
Statistics, Line Plot, Bar Chart - exploring the value of V1-V28 and amount across the classes
Logistic Regression Learner, Predictor, Scorer - training a new model using V1-V28 (null hypothesis rejected)
To use this workflow in KNIME, download it from the below URL and open it in KNIME:
Download WorkflowDeploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.
Try NodePit Runner!