Icon

Telecom Churn Prediction

This workflow is an example of how to build a basic PMML model for a churn prediction using a Decision Tree algorithm.

URL: Build a workflow https://www.knime.com/getting-started
URL: KNIME Quickstart Guide https://docs.knime.com/latest/analytics_platform_quickstart_guide/index.html

Data Reading



Pre-processing


Train a Model


Churn = 0 customer remained with contract

Churn = 1 customer quit contract

training set

test set

Visualize and Score


Component
Match original vs. predicted Churn values
Scorer
Save model in PMML format
PMML Writer
Training set 80% vs. Test set 20%
Table Partitioner
Apply the trained model to predict Churn
Decision Tree Predictor
Column Resorter
Reading ContractData.csv
CSV Reader
Join the contract data and the behavioral data
Joiner
Target Class = Churn
Decision Tree Learner
Reading CallsData.xls
Excel Reader
Area code and churn 0/1 are converted to String.
Number to String

Nodes

Extensions

Links