Siebel Smart Answer Administration Guide > Configuration and Implementation of Siebel Smart Answer >

Prerequisite Tasks

Before you can use Smart Answer, you must perform some required tasks. You do not perform these prerequisite tasks in Siebel Business Applications. You perform some of these tasks in Banter Workbench.

To complete the prerequisite tasks for setup, follow the instructions in the Banter documentation. Also read these guides to familiarize yourself with the Smart Answer terminology. For information about how to locate Banter documentation, see Banter Documentation.

Perform the following tasks in the sequence shown:

1. Gather and analyze the data that you want Smart Answer Manager to categorize. This data is the source data for the corpus file that you use to generate the KB file.
2. Determine the categories to use in your KB for classification and training.

   Choosing categories is one of the most important steps in building an initial KB file. A category is a classification that you assign to each type of inquiry or request that you receive from customers. You define your own categories based on your business requirements during the Smart Answer planning process. Categories might represent actual content or intent, or might identify another attribute of the text, such as the name of the sender. Examples of categories are product complaint, service complaint, and order status.

   Smart Answer Manager returns category names and their scores to your Siebel application. Siebel Email Response can use this information to automatically respond to emails, suggest categories of response to agents, route emails, and perform other workflow tasks. Smart Answer Manager stores all the data that it needs to correctly identify intents and concepts in the KB, including information about the individual categories. This information is the foundation for the concept models.

   Smart Answer for customer or employee applications uses the content items associated with the category to display various content types like solutions, decision issues, literature items, and resolution items.

3. Install Banter Workbench. You need Banter Workbench 5.5s to create and train the KB file. This tool replaces both the KB Editor and the RME Jumpstart used in previous releases of Smart Answer. The installation files are on the Windows Client Third-Party CD-ROM in the \Thirdpty\Banter folder.
   1. Create a folder named Smart Answer Tools on your local machine, and install Banter Workbench 5.5s in that folder.

      Banter Workbench 5.5s allows you to create or import a corpus of texts, train and test a KB file, and analyze the performance of the trained KB file.

      Run the Setup.exe file to install the software in your local folder. This program also contains a Knowledge Base Editor which helps you to create the structure of the KB file. The Knowledge Base Editor has two parts, as follows:

      • Dictionaries and methods that are used to analyze a message.
      • Concept models that are used to detect the intent of a message.

        For simplicity, use a flat KB file at first, unless you are creating a multilingual KB.

   2. If your KB file grows to contain a large number of categories, you might need to convert it to a hierarchical KB file to increase the accuracy of Smart Answer feedback.

      You can use Banter Workbench to convert a flat KB file into a hierarchical KB file. The conversion is a manual process. You use the KB Editor to add branches for each group of categories.

      If you change the structure of a KB file, you lose all the learning and feedback that has been accumulated. You must train and test the KB file again. Keep an updated corpus of emails for retraining purposes.

      Note the following points relevant to hierarchical KBs:

      • To use a multilingual KB, you must create a hierarchical KB. For more information about multilingual KBs, see the Banter Workbench User's Guide.
      • When you create a hierarchical KB, you must assign branches and categories unique names. For example, you cannot create a branch called Order Status and a category called Order Status. If you assign the same name to a branch and a category, the category name overwrites the branch name and only the category appears in the Siebel catalog file.
      • The maximum length for category names is 100 characters.
4. Start a new project in Banter Workbench. For information about how to start a new project, see Banter Workbench User's Guide.

   NOTE:  Give the project the same name as the KB file for consistency. The project becomes the KB file which you import to your catalog, and which resides in the Siebel File System.

5. Define the corpus.

   When you define the corpus, you specify which email fields you want to train your KB on. By default, Siebel Email Response sends the Body field and the Subject field to the categorization engine. You can add any number of fields that you want to train the KB on as long as those fields exist in your corpus.

   When you set up the NLP usage options in Banter Workbench, associate NLP options with all the fields that you plan to send to the categorization engine in Smart Answer. Perform this action if you want to make sure that the confidence levels match when generated by the following methods:

   • When you select the option called Create using all selected, analyze using all selected.
   • When the system reads the text of an incoming email in Siebel Email Response.
6. Create and edit your corpus file.

   A corpus contains your texts for training assigned to categories. A corpus can be created in a number of formats. For more information about the supported formats, see the Banter Workbench User's Guide.