Text classifier
Train the model to label text columns from examples.
Train the model to label text columns from examples.
This tool looks at all the available text classifier tools and picks the best one.
Note
The speed and quality slider is more of a spectrum. This setting determines the number of machine learning models considered. Models that do not support the training data are automatically excluded.
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 when you want to predict values.
Configuration
Use the following configuration options to configure the Text classifier tool.
Go to the Pipes module from the side navigation bar.
On the Pipes tab, find the pipe you want to work with. Click the pipe to open. For more information about pipes, see the Creating a pipe documentation.
In your Pipe builder, add your data source.
On the canvas toolbar, click
+ Tool.
In the Tools modal search bar, type Text classifier.
Tip
You can also find the Text classifier tool in the Learn section.
Connect the tool to your data set.
In the configuration pane, enter the following information:
Table 72. Text classifier tool configurationField
Description
Text classifier type
Choose which text classifier type to use:
Automatic
CNN Text Classifier
BOW Text Classifier
Ensemble Text Classifier
Text column
Select the text column to use the Text classifier for the prediction and probability.
Target column
Select the target column to use the Text classifier for the prediction and probability.
Speed versus Quality slider
Use the slider to indicate if you want speed versus quality when the Text classifier is working.