Create a predictive model

When you create a predictive model, you need to specify the document population by selecting a saved search or a folder.

Prerequisites: Make sure the saved search or folder you use to specify the document population includes at least one document coded as positive and one coded as negative to run the model’s initial training. Adding more coded examples improves the model’s accuracy.

To create a predictive model, complete the following procedure.

1. In the Main Menu, select Review.

2. Click and select Create Predictive Model.

3. In the Create Predictive Model dialog box, do the following.

   1. In Model Name, enter a name for the predictive model.

   2. In Population, do one of the following.

      • Select Saved Search, and then choose a saved search from the list.

      • Select Folder, and then choose a folder from the list.

   3. In Tag Settings, do the following.

      1. In Single Choice Field, select a field.

         • You must select the single-choice field and the positive and negative values used to code the documents when preparing the document set for the predictive model.

      2. In Positive Value, select the positive field value.

      3. In Negative Value, select the negative field value.

   4. Click Create.