Classifier Model Reference Data
A classifier model contains a reference data column that can include sentences, paragraphs, or pages of text. The reference data represents the different types of text input that a Classifier transformation can read in a mapping. When you create a model, verify that the reference data includes the types of text that you expect to find when you run the mapping.
You can use the mapping source data to create a classifier model. Select a sample of the source data and copy the data sample to the model.
Consider the following rules and guidelines when you work with classifier model reference data:
- •A reference data field can be of any length. You can enter pages of text into each data field.
- •You import reference data from a data object.
- •You cannot edit reference data values. However, you can delete a data row.
Adding Data and Label Values to a Classifier Model
Use a data source to append data to a classifier model. You can add data values and label values.
1. Open the content set that contains the model.
2. Select the model name and click Edit.
3. Click Append Data.
The Classifier Model wizard opens.
4. Browse the Model repository and select the data object that you want to use. Click Next.
Note: Do not select a social media data object as a data source.
5. Review the columns on the data object, and select a column to add as a data column or label column for the model. You can add a reference data column and a label column in the same operation.
- - To use a data source column as the reference data column in the model, select the column name and click Data.
You can select multiple data columns. The classifier model merges the contents of the columns you select into a single column.
- - To use a data source column as the label column for the model, select the column name and click Label.
Click Next.
6. Select the number of rows to copy from the data source.
Select all rows, or enter the number of rows to copy. If you enter a number, the model counts the rows from the start of the data set.
7. Click Finish and save the model.
After you append the data, verify that the data rows you added include label values.
Deleting Data Values from a Classifier Model
You can delete data values in the default view and in the detailed view of a classifier model. To delete all data values, use the default view.
1. Open the classifier model in the Developer tool. To open the model, select the model name in the content set and click Edit.
2. Select the row that contains the data you want to delete.
You can select a single row, multiple rows, or all rows.
3. Click Delete.