Icon

Deploying a Churn Prediction Model

Deploying the churn predictor
Churn Prediction: DeploymentTask Deploy a previously trained model to predict the churn for new customer data. Using PMML we only need 4 nodes for the whole deployment workflow. PMML is transparent to the model type, be it a neural network or adecision tree, the PMML Predictor node understands everything.The workflow is also available in the EXAMPLES server under 50_Applications/_18_Churn_PredictionA use case is described at URL: https://www.knime.org/knime-applications/churn-predictionImplementation is described in this KNIMETV video "Building a basic Model for Churn Prediction with KNIME" https://youtu.be/RHsO10q7e2Y Data & Model ReadingIn the file newdata.csv in TheData/Customers folder, reallife data are simulated. The file contains behavioral andcontract data, without churn information, for one newcustomer only.The previously trained model is read. It can be any kind ofmodel saved in PMML format. Apply the ModelPMML Predictor is thePMML interpreter node Display Churn Score in a ReportExport results with churn prediction to report. Try this:Create your own report:1) Execute the entire workflow 2) Click the “Open the report” button inthe toolbar (make sure that the KNIMEReport Designer extension is installed. Ifnot, go to File, then Install KNIMEExtensions/KNIME Report Designer)3) Select "Run" from the reportenvironment 4) Choose your preferred format to seethe final report.Note: If you have changed the KNIMEWorkbench, click View/ResetPerspective to see all panels in theReport Designer. Read newdata.csvRead previously trained modelProduce ScoreCardfor reportApply PMML model at input port to input data File Reader PMML Reader Data to Report PMML Predictor Churn Prediction: DeploymentTask Deploy a previously trained model to predict the churn for new customer data. Using PMML we only need 4 nodes for the whole deployment workflow. PMML is transparent to the model type, be it a neural network or adecision tree, the PMML Predictor node understands everything.The workflow is also available in the EXAMPLES server under 50_Applications/_18_Churn_PredictionA use case is described at URL: https://www.knime.org/knime-applications/churn-predictionImplementation is described in this KNIMETV video "Building a basic Model for Churn Prediction with KNIME" https://youtu.be/RHsO10q7e2Y Data & Model ReadingIn the file newdata.csv in TheData/Customers folder, reallife data are simulated. The file contains behavioral andcontract data, without churn information, for one newcustomer only.The previously trained model is read. It can be any kind ofmodel saved in PMML format. Apply the ModelPMML Predictor is thePMML interpreter node Display Churn Score in a ReportExport results with churn prediction to report. Try this:Create your own report:1) Execute the entire workflow 2) Click the “Open the report” button inthe toolbar (make sure that the KNIMEReport Designer extension is installed. Ifnot, go to File, then Install KNIMEExtensions/KNIME Report Designer)3) Select "Run" from the reportenvironment 4) Choose your preferred format to seethe final report.Note: If you have changed the KNIMEWorkbench, click View/ResetPerspective to see all panels in theReport Designer. Read newdata.csvRead previously trained modelProduce ScoreCardfor reportApply PMML model at input port to input data File Reader PMML Reader Data to Report PMML Predictor

Nodes

Extensions

Links

Download

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

Download Workflow
On NodePit since: 2019-12-14
Last update: 2024-04-25
Created with KNIME version: v4.0.0

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

Try NodePit Runner!