Objective
The objective of this workshop is to build a working machine learning (ML) model in 30 minutes without coding using IBM SPSS Modeler. You’ll build a model that will help to predict the customer churn (customers leaving the company) for a telecommunications company (telco).
In order to run this workshop you must have IBM SPSS Modeler intalled on your computer (PC or Mac). If you don't have a copy of IBM SPSS Modeler, check below
Obtaining a copy of IBM SPSS Modeler
- Get a 30 day free IBM SPSS Modeler trial version from the IBM website - via IBM Watson Studio Desktop trial
- IBM SPSS Modeler student edition for students and professors via On the Hub
- Use the IBM Digital Technical Enablement (DTE) IBM SPSS Modeler virtual environment
This repository contains the following assets for running the workshop
Workshop setup
- Open IBM SPSS Modeler om your Mac or PC (or use the virtual environment mentioned above)
- Download the workbook and the data needed for the workshop
- Open the workbook (PDF) and follow the steps to complete the workshop
Workshop
The workhop consists of the following steps which are all outlined in the workbook.
- Connect, merge and augment the data sources
- Exploratory Data Analyses (EDA)
- Build the ML model
- Evaluate and visualize ML model performance
- Make predictions on a new data set
Once you have completed the workshop the end result will look like the screenshot below.
