Icon

05. Modelling Workflow with Integrated Deployment - solution

Modelling Workflow with Integrated Deployment

This workflow shows a simple example of churn prediction with interactive threshold optimization. This workflow is used for showcasing Integrated Deployment introduce in KNIME 4.2. You can see how the trained model is automatically captured (via Capture Workflow nodes) and deployed (via a Workflow Writer node).

Train and Optimize Pre-processing - Join contract data and behavioral data - Convert Churn values to String to be used as class in upcoming classification - Reserve 80% of the rows for model training and remaining for model testing test set Reading and Blending - contract data + churn - behavioral (calls) data Score Evaluate predictions based on confusion matrix views. train set Capture for Deployment : Scoring - Missing value imputation modelling - Optimize Random Forest parameters - Optimize threshold Binary Classification - Train model with optimized parameters - Capture branch to deploy Capture for Deployment : Data Preparation Churn = 0 : current subscriptions Churn = 1 : cancelled subscriptions Deploy Automatically write the deployment workflow train: 80% test: 20%ReadingContractData.csvJoin the contract data and the behavioral dataArea code and churn are converted to String. optimized modeloptimize thresholdto max accuracyapply newthresholdoptimized thresholddefault 50% thresholdDB Table Selector DB Connector Partitioning File Reader(deprecated) Joiner (deprecated) Number To String Random ForestLearner Random ForestPredictor Missing Value Missing Value(Apply) Binary ClassificationInspector Rule Engine CaptureWorkflow Start CaptureWorkflow End CaptureWorkflow Start CaptureWorkflow End Scorer (JavaScript) Workflow Combiner Scorer (JavaScript) DB Reader Database URL andCredentials Domain Calculator ParameterOptimization Workflow Writer Table Rowto Variable Train and Optimize Pre-processing - Join contract data and behavioral data - Convert Churn values to String to be used as class in upcoming classification - Reserve 80% of the rows for model training and remaining for model testing test set Reading and Blending - contract data + churn - behavioral (calls) data Score Evaluate predictions based on confusion matrix views. train set Capture for Deployment : Scoring - Missing value imputation modelling - Optimize Random Forest parameters - Optimize threshold Binary Classification - Train model with optimized parameters - Capture branch to deploy Capture for Deployment : Data Preparation Churn = 0 : current subscriptions Churn = 1 : cancelled subscriptions Deploy Automatically write the deployment workflow train: 80% test: 20%ReadingContractData.csvJoin the contract data and the behavioral dataArea code and churn are converted to String. optimized modeloptimize thresholdto max accuracyapply newthresholdoptimized thresholddefault 50% thresholdDB Table Selector DB Connector Partitioning File Reader(deprecated) Joiner (deprecated) Number To String Random ForestLearner Random ForestPredictor Missing Value Missing Value(Apply) Binary ClassificationInspector Rule Engine CaptureWorkflow Start CaptureWorkflow End CaptureWorkflow Start CaptureWorkflow End Scorer (JavaScript) Workflow Combiner Scorer (JavaScript) DB Reader Database URL andCredentials Domain Calculator ParameterOptimization Workflow Writer Table Rowto Variable

Nodes

Extensions

Links