This documentation supports the 22.1 version of BMC Helix Innovation Studio.To view an earlier version, select the version from the Product version menu.

Creating search filters for BMC Helix Virtual Agent


BMC Helix Knowledge Management by ComAround supports filters to narrow down search results in BMC Helix Virtual Agent based on predefined conditions. These filters use article tags as filtering criteria. A single filter can combine multiple article tags by using the AND or OR operators. When users search for knowledge articles in a chatbot that utilizes a specific filter, this chatbot returns only the articles that include a tag or a combination of tags defined in the filter. 

As an administrator of BMC Helix Innovation Studio, you can create search filters for BMC Helix Virtual Agent. After you create a search filter in BMC Helix Innovation Studio, you must enable it for the required chatbot instance in BMC Helix Virtual Agent. For more information, see Defining a knowledge search provider for chatbot.

Important

You can create and use search filters only for BMC Helix Virtual Agent. This functionality is currently not supported for other BMC applications integrated with BMC Helix Knowledge Management by ComAround. 

Scenario

Scenario

In BMC Helix Virtual AgentApex Global has created multiple specialized chatbots to address employees' queries about different lines of business. The company wants each chatbot to return only the knowledge articles that relate to a specific line of business. Apex Global uses BMC Helix Knowledge Management by ComAround as a search provider for BMC Helix Virtual Agent. Therefore, the company can create filters to limit search results in chatbots.

As an administrator at Apex Global, Jordan must create a search filter for the HR chatbot. The following image shows how Jordan creates this filter and how Apex Global and its employees benefit from using it.  

Scenario_and_benefits.png

To create a search filter

  1. Log in to BMC Helix Innovation Studio and click Administration
  2. Select Knowledge management > Search filters
  3. On the Search filters page, click New search filter and specify the following fields:

    Field

    Action

    Name

    Enter the name for a search filter.

    Relationship

    Select one of the following logical operators to define the relationship between filtering criteria: 

    • AND—Returns articles that meet all filtering criteria. For example, Line of Business=HR and Operational Tier 1=People.
    • OR—Returns articles that meet any filtering criterion. For example, Line of Business=Office Facilities or Knowledge Set=Office.

    Important: You cannot combine AND and OR logical operators in a single filter. You can use only one operator to define the relationship between all selected filtering criteria. For example, you cannot define the following filtering expression: Company=Apex Global and Line of Business=Office Facilities or Knowledge Set=Office.

    Tags

    Select one or multiple predefined article tags to be used as filtering criteria.

    Important: Only existing article tags are displayed in the list.

    Tip: To see the list of predefined tags, click on a text input area of the Tags field. Alternatively, you can enter the tag name. When you begin entering the tag name, a list of suggested selections appears.

    To remove a selected tag, hover over the tag and click the Remove remove_icon.png icon. 

    The following image shows a sample search filter. In the example scenario described in this topic, Jordan creates the same filter for the HR chatbot. 
    search_filters.png

  4. Click Save

You have created the search filter. This filter is displayed on the Search filters page. To modify this or other created filters, open the filter by clicking its name. 

Important

When you edit existing filters, the Save button becomes available only after you make any changes to the filter. 

Where to go from here

After you create the search filter, you must enable it for the required chatbot instance in BMC Helix Virtual Agent. For more information, see Defining a knowledge search provider for chatbot.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*