In our recent post Capture the ROI of predictive analytics with 3 projects, we highlighted some powerful quick wins you can achieve using SPSS Modeler from IBM. With the right predictive analytics tools you can solve key business problems such as reducing churn and increasing cross-sell and up-sell.
Using SPSS Modeler for predictive analytics and machine learning
Want a closer look at how SPSS Modeler does its magic? Let’s pop the hood and take a quick tour. We’ll use a project for predicting attrition and customer churn. We’ve found this is one of the best ways to learn SPSS Modeler.
Below you’ll see an example of an analytical stream that predicts churn. A stream is simply a visual representation of a predictive analytics model. We’re using historical customer data from two databases. The data is prepared so that each row represents one customer, and their associated activity.
The building blocks of the Modeler stream are called “nodes,” and at each node, SPSS Modeler performs operations on the data. The data contains multiple variables, including that variable we want to predict, “Churn.”
Visualizing a machine learning model to predict attrition and customer churn
Here’s a breakdown of the predictive model nodes in the stream for predicting customer churn (from left to right):
- CRM database and Transactional data nodes: Source nodes reflect where we’re accessing data; Modeler can use a variety of databases such as SQL, SAS, Excel, XML, .csv, etc.
- Merge node: Joins the two data sets together for modeling, with one record for each customer.
- Type node: Labels variables as continuous (numeric) or categorical, and identifies the target variable. In this case the target variable is “Churn”, with two values: “Churn” or “No Churn”.
- Partition node: Partitions the data into two sets. A training set we use to build the model and a testing set we’ll use to evaluate the accuracy of our model.
- CHAID Churn node (pentagon): CHAID (CHi-squared Automatic Interaction Detection) is the decision tree classification algorithm we use to build a model to predict the Churn variable. Modeler offers a wide range of modeling algorithms, including neural networks, many of them automated.
- CHAID Churn model node (gold nugget): This is the decision tree model created by the CHAID algorithm to predict Churn. It assigns each record in the data a prediction of “Churn” or “No Churn” and a confidence score for each prediction.
- Evaluation node: This node evaluates the accuracy of our model on the training and testing data sets using a cumulative gains curve or lift curve.
- Database node: We can use our model to score new data and write it into a database.
Once you’ve got predictive data in your database, you can translate it into actionable insights for other teams. In the case of preventing churn, your Marketing team is a key stakeholder. Using the model, they could identify which customers are most likely to leave and engage them with targeted incentives to stay. While we are using SPSS Modeler for this example, you would follow these same general steps using another predictive analytics tool, or a programing framework like Python.
How Beyond the Arc can help with SPSS modeler and Machine Learning & AI
Our team includes experts in data science, passionate about making decisions and taking action based on data. We specialize in using machine learning to deliver predictive analytics ROI.
Take advantage of a complete predictive analytics platform — as an authorized reseller and certified IBM Business Partner, Beyond the Arc can provide IBM SPSS Modeler software, training, and hands-on model building and consulting.
Interested in learning more about how to deploy AI? Let’s start a conversation.