Setting up chatbots for your line of business

In addition to the default chatbot, BMC Helix Virtual Agent enables the administrators and application business analysts to set up specialized chatbots in an organization. This capability enables you to create specialized chatbots for each individual team or department, and train these chatbots so that they are experts in specific areas without any risk of overlapping. You can create a single or multiple specialized chatbots in a BMC Helix Innovation Suite environment.

For example, you can create an HR Requests chatbot that addresses HR queries only and an IT Helpdesk chatbot that addresses IT queries only. These specialized chatbots help to ensure that the employee queries are quickly addressed.

Before you begin

• Ensure that you have subscribed to the appropriate service license to configure the chatbot.

• Ensure that you have performed the following tasks in the specified products:

  Product	Task
  BMC Helix Innovation Studio	Created and deployed applications for which you want to set up the chatbot.
  IBM Watson Assistant	If you have subscribed to IBM Watson Assistant Plus or IBM Watson Assistant Premium plan, as an administrator, provide skill-level permissions to your users. For more information, see To provide skill-level permissions in IBM Watson Assistant service for chatbot.
  IBM Watson Assistant	You have the Identity and Access Management (IAM) API key of the IBM Watson Assistant instance in which you have created the chatbot Skill. You can get the IAM API Key by logging in to IBM Cloud. Copy the IAM API Key from this screen
  IBM Watson Assistant	You have the IBM Watson Assistant Skill ID of the reference Skill that is provided with BMC Helix Virtual Agent or custom IBM Watson Assistant that you want to use.  You can get the Skill ID by logging in to the IBM Watson tool and selecting the appropriate skill.  Copy the Skill ID from this screen
  (If you want to use custom or localized Skill) IBM Watson Assistant	The Skill was created in the IBM Watson Assistant tool.  The chat actions for BMC Helix Business Workflows and BMC Helix Digital Workplace Advanced are configured in the BMC Helix Virtual Agent dialog. For localization, the chat actions must be configured for each skill defined for a locale. For JSON example actions, see Example JSON of BMC Chatbot dialog responses.
  (If you want to use real-time translation to localize chatbot conversations) BMC Helix Innovation Studio	Enable real-time translation for chatbots. For more information, see Enabling-a-real-time-translation-provider-for-chatbots.
  (To use disambiguation) IBM Watson Assistant	By default, disambiguation is always enabled. To learn how to edit the disambiguation configuration, see Disambiguation example in IBM documentation.

Process for setting up chatbots for your line of business

The following image gives an overview of the tasks involved in setting up chatbots:

To create a chatbot

1. Log in to BMC Helix Innovation Studio and navigate to the Workspace tab.
2. Select the application for which you want to create a chatbot. You can select the application based on access level for users:
   • To provide access to all users, create a chatbot in the BMC Chatbot application.
   • To provide access to a specific group of users, create a chatbot in the application that only the user group has access to.
3. On the Chatbots tab, click New and enter a unique name for the chatbot. 
   For example, IT Helpdesk.
4. Click Save.

The newly-created chatbot is displayed in the Chatbots tab, as shown in the following image:

To configure the Skill ID for a chatbot

In BMC Helix Virtual Agent, you must provide the Skill ID, so that the BMC Helix Cognitive Automation can connect to your IBM Watson Assistant skill. Additionally, if you want to use localization you must provide the Skill ID of IBM Watson Assistant skill for each locale.

1. Log in to BMC Helix Innovation Studio and navigate to the Workspace tab.
2. Select the application for which you created a chatbot.
3. From the Chatbots tab, select the newly created chatbot.

4. In the Chatbot Skills tab, fill out the following fields:

   UI field	Description
   API Key tab	Enter the Identity and Access Management (IAM) API key of your IBM Watson Assistant service that you noted down earlier.
   Workspace ID	Enter the ID of the IBM Watson Assistant that you want to use.
   Locale	View the name of the locale, which is automatically populated according to the Skill ID that you entered earlier.
   Chatbot Display Name	Enter a name for BMC Helix Virtual Agent that will be displayed to users.
   Welcome Message	Enter the message that you want to be displayed when the user first opens the chatbot. When you use localization, you can enter the localized welcome message for each locale that you want to use.
   Request Clarification Prompt	Enter the phrase to be displayed when the chatbot is unable to identify the user's exact intent. Example: Did you mean...
   No Relevant Options Label	Enter the phrase to be displayed when none of the options provided by the chatbot are relevant for the user. Default: None of these Note: If you keep this field blank, this option is not displayed in the chat conversation.

   The following image shows the fields in the Chatbot Skills tab:
   

5. Click Save. 
   After saving, disambiguation is enabled in IBM Watson Assistant.

To view the chatbot and verify the chatbot settings, click Launch Bot. 

To add service option buttons in the welcome message

1. Log in to  BMC Helix Innovation Studio and navigate to the Workspace tab.
2. Select BMC Chatbot. 
3. On the Chatbots tab, select BMC Chatbot > Chatbot Skills.
4. On the Chatbot Skills tab, click Add/Edit Message.
5. In the Welcome Message box, add a message to welcome the end users. The welcome message can be typed in multiple lines.

1. From the Intent list, select a target intent.

2. From the Button Label list, select a button label to display. 

   Important

   These labels are the utterances in the IBM Watson Intent settings. If you want to add a different label, please visit the IBM Watson page and add that label as an utterance. For more information, see Configuring-IBM-Watson-Assistant-to-work-with-BMC-Helix-Virtual-Agent.

3. Click Insert. 
   The button appears in the Option Buttons section. You can add multiple buttons.
   
4. (Optional) To remove a button, select it and press Delete.
5. (Optional) To re-position a button, select and drag it to a different position. 

6. Click Save.
   The following image shows how the service option buttons appear on the Chatbot UI.
   
   
   

   Important

   If the Skill ID is changed, you will get the following error: 

   Buttons must be linked to intents in the Workspace. 

   In such a scenario, update the target intents for each button so that they point to the intents of the correct Skill ID.

To configure a chatbot to work with a channel

A chatbot is an omni-channel application that can work with several communication channels. When you create a new chatbot, by default, it is configured to work with the BMC Helix Virtual Agent web UI. You can configure the chatbot to work with different user communication channels such as Slack, Microsoft Office 365 (Teams), Skype for Business on-premises, SMS, and WhatsApp (by using Twilio). Although this is an optional task, it is recommended that you leverage this capability of chatbot. 

For example, you might want to provide SMS chatbot conversation facility to your application users. To provide the SMS facility, you must configure your chatbot to work with Twilio so that your application users can interact with a chatbot by sending text messages through their mobile devices.

BMC Helix Virtual Agent supports the following communication channels:

To change the branding of a chatbot

You can change the branding of a chatbot by uploading your organization's logo, header color, header text, and so on. You can also customize the chatbot animation when it loads for the first time. Although this is an optional task, it is recommended that you leverage this capability of chatbot. The branding applies only to the chatbot web UI and not to other chatbot channels such as Office 365, Skype for Business on-premises, Slack, and Twilio. 

To learn how to customize a chatbot and the best practices for file size and pixel size, see Branding-BMC-Helix-Virtual-Agent-for-your-company.

To configure the dynamic options and real-time translation for chatbot

After you create a chatbot, you can configure the dynamic options that are returned to a user in a chatbot conversation based on the user query. For example, you can configure the maximum number of knowledge articles, maximum number of answer options displayed per question, and so on. Although this is an optional task, it is recommended that you leverage this capability of the chatbot. 

1. Log in to BMC Helix Innovation Studio and navigate to the Workspace tab.
2. Select the application for which you created a chatbot.
3. On the Chatbots tab, select the newly created chatbot.

4. Click General Settings and enter the details for the following settings:

   Setting name	Action
   Maximum Number of Requests for Status to Return to User	Specify the maximum number of service requests related to the user query. The default value is 5.
   Maximum Number of Knowledge Articles to Return to User	Specify the maximum number of knowledge articles that are relevant to the user query. The default value is 5.
   Maximum Number of Options/Buttons to Display Per Question on Web Channel	Specify the number of options or buttons that must be displayed. End users click the options or buttons to respond to a question by the chatbot. In the Show More menu, you can view the options or buttons in addition to the number that you have configured. The default value is 3.
   Maximum Number of Passage Characters	Specify the number of characters that you want to include in the knowledge article paragraph displayed during cognitive search. The default value is 400. Best practice We recommend that for best results, you must set the maximum number of passage characters to a value over 200.
   Maximum number of Dynamic Answer Options to Display on Twilio	Specify the number of options or buttons for dynamic answers that must be displayed on Twilio. The default value is 5.
   Maximum Number of Assets to Return to User	Specify the number of assets that should be returned to the user. The default value is 16.
   Maximum Number of Business Services or Resource Pools to Return to User	Enter the maximum number of services or resource pools that you want to show to the user in a chat. Important: The default value of this field is 10 and the valid range is 5-20.
   Maximum Number of Resources or VMs in Recommendation to Return to User	Enter the maximum number of resources or virtual machines that you want to show in a recommendation. Important: The default value of this field is 5 and the valid range is 5-20.
   Default Locale	Select a default locale. The default locale can be used as the preferred language for chatbot conversations if the end user's language preference is not supported by the Chatbot Skills or real-time translation is not enabled. The default locale is also used for translation if the real-time translation is enabled. Important: You cannot remove a language skill that is set as the Default locale. Before removing it, you must add another language skill and set that as the default.
   Real-time Translation	If you have configured a real-time translation provider, turn this toggle key to On. Best practice If you have set the preferred user locale for all users of a tenant, we recommend that you enable real-time translation for all the chatbots. Otherwise, end users might get an inconsistent localization experience.
   Rating Scale Presentation Order	Specify the order in which you want to display the rating scale options. The default order is from ascending to descending.
   User Name Presentation	Specify how you want the chatbot to address the end users. By default, the chatbot addresses the user by first name, middle initial, and last name. For example, Britney V Smith. You can also display only the first name or select None if you don't want to display the user name in the welcome message.Important: If you select to display the first name only and if the user has specified a nick name, the nick name is displayed.

5. Click Save. 
6. (Optional) To revert to the default value, click Reset to Default and then click Save.

In addition to these settings, you can configure search and live chat for the chatbot. 

To configure the chat session idle timeout

You configure the idle timeout period after within which the chat session closes automatically. The default timeout period is 15 minutes. 

1. Log in to BMC Helix Innovation Studio and navigate to the Administration tab. 
2. Select Configure My Server >  Cognitive Service.
3. From the Configure list, select Chatbot.

4. In the Chatbot section, enter the value in the following field, and click Save.

   Field	Description
   Chat Session Idle Time	Enter the time in seconds after which the chat session closes automatically if the end user is not chatting. Default time: 900 seconds Valid range: 300 to 1200 seconds

To view all the chatbots in the BMC Helix Innovation Suite environment

Administrators can view all the chatbots that are created for all applications in the BMC Helix Innovation Suite environment.

1. Log in to BMC Helix Innovation Studio and navigate to the Administration tab.
2. Select Cognitive and chatbot > Chatbots. 
   The Chatbots list displays all the chatbots that are configured in the BMC Helix Innovation Suite environment. 
   For example, the following image displays the default chatbot, BMC Chatbot and a custom chatbot. 
   
   
   
3. (Optional) From the Chatbots list, select the chatbot name that you want to configure.
   The configuration window for the selected chatbot is displayed. For example, if you select the IT Helpdesk chatbot, the IT Helpdesk chatbot configuration window is displayed.

Related topics

Utilizing-multiple-chatbots-for-each-line-of-business

Importing-chat-enabled-services-from-BMC-Helix-Digital-Workplace-Advanced

BMC-Helix-Virtual-Agent-usage-reports