Train the model to predict a column with a set of values.

The Predictor first determines if the problem looks like a classification or a regression. If it does, then this tool considers all classification modeling tools and regression modeling tools. Varicent ELT uses the tool with the best score to make a prediction.

This tool adds two columns to your data: a prediction and a probability. It's the model's certainty about the prediction.

When you run the tool, the data is automatically split: 80% of the data is used for training. The remaining 20% is used for testing. Each model being considered is trained and evaluated to select the one with the best score. This is done 5 times to predict the test values (the 20% of your data). The final score is the average of all 5 scores.

When to use this tool

Use this when you want to make a prediction, but you're unfamiliar with classification and regression modelling tools. After you're comfortable with the difference between those tools, switch to using the Regressor or Classifier tool.

Figure 9. Predictor tool example

Configuration

Use the following configuration options to configure the Predictor tool.

Configuring the Predictor tool

1. Go to the Pipes module from the side navigation bar.

2. From the Pipes tab, click an existing pipe to open, or create a new pipe. To create a new pipe, read the Creating a pipe documentation.

3. In your pipe, add your data source.

4. Click + Tool.

5. In the Tools modal, search for Predictor. Click + Add Tool.

   Tip

   You can also find the Predictor tool in the Learn section.

6. Connect the tool to your dataset.

7. In the configuration pane, enter the following information:

   Table 73. Predictor tool configuration

   Field	Description
   Target column	Select the column that you want to use the Predictor on.
   Advanced section
   Speed vs Quality slider	Use the slider to indicate if you want speed versus quality when the predictor is working.
   Exclude columns	Select the column(s) that you want to exclude from the prediction.
   Smart exclude	Select this option if you want to have Smart Exclude identify and automatically exclude columns that don’t help predict the target column after you build.

   
   
   

The Prediction and Probability columns appear with the predication and the probability of the prediction.

Also, check the Tool tab for more detail on the model.

Usage example

For example, you want to predict customer churn based on historical data. The dataset that you're working with includes various attributes about customers and whether they have already churned or not.

Using the dataset, the Churned column is the one that we want to use to predict. We want the tool to analyze the data and model dependencies between the Age, Tenure, LastTransactionAmount, and the other columns to predict the churn for a customer.

The original dataset is augmented with predicted Churned values and associated probabilities, helping the company forecast customer behaviour trends. This gives you the opportunity to take preemptive actions such as a targeted retention campaign for customers predicted to churn.