For the virtual assistant to recognize a dialog flow, follow these steps:
1. Here you can add conditions to recognize the dialog that you’ll be creating.
2. Map the Dialog with Intent, Entity, or Variable. This module also consists of options to add AND/OR conditions by clicking on the Add More button. With this functionality, you can make the Dialog based on more than one condition.
3. To specify the type of condition, use the Select dropdown on the Recognize Dialog pop up. And choose from: Intent, Entity and Variable.
4. Let’s say you select Intent, you’ll then be presented with another dropdown to select the Intent from the list of existing Intents (you created). You can create a new Intent on the Recognize Dialog pop up itself by typing in the name of the Intent you want to create and clicking on Create New that appears after typing in the Intent name.
5. You can even specify the examples for this new Intent on the Recognize Dialog pop up itself by typing in the examples on the pop up that appears after you specify the Intent name.
6. After selecting the Intent, you’ll be presented with another dropdown showing the intent is recognized.
7. You can add another Intent or an Entity or Condition using the And/Or logic operators. When you select Entity as the condition type, you’ll be presented with the following fields under it.
8. Under Entity, you can either select an already created entity or a pre-existing system entity and select a rule such as ‘is recognized’, ‘value is’ and ‘value is not’.
Condition: A condition is used for the assistant to recognize the user or some of his/her traits during interaction. Some common examples of this can be recognizing a user based on his/her time zone, name, or even phone number.
For example, if you want to greet a user by their name, you can do so by setting the condition with the help of ‘context variables’ which will recognize the username and greet them. We’ll get into more details of how you can use context variables as we progress.
Operators vary depending on whether you’ve decided to go with an intent, entity or a variable.
The following logic will be applied by default under Intent:
Is Recognized — Checks whether any of the examples under the chosen intent is present in the user’s query.
Ex: If you create an Intent named Unlock_Account with examples as:
- Unlock account
- Account locked
- Unable to access my account
If user types My account is locked. Then bot will identify it as Unlock Account intent and checks for the dialogs that use Unlock Account Intent with is recognized as the condition. In this scenario, Unlock Account dialog will be recognized and the VA responds to the user with the dialog conversation flow.
The following are the operators that come under Entity on the Recognize Dialog pop up:
value is — Checks whether a specific entity “value” synonym or regex of the chosen entity is present in the user’s query
value is not — Checks whether a specific entity “value” synonym or regex of the chosen entity is not present in the user’s query
is recognized —Checks whether any of the entity values synonyms or regex under the chosen entity is present in the user’s query
The following operators come under Condition (Context Variable) on the Recognize Dialog pop up. Depending on your selection in the previous field, the operators change accordingly. For some operators, a ‘field’ will be displayed to select the ‘variable’ with which the input variable must be compared to check for selected condition.
Equals (Text) — Checks whether the user’s query contains the specified text from the chosen variable.
Does Not Equals (Text) — Checks whether the user’s query doesn’t contain the specified text from the chosen variable.
Is Present — Checks whether the user’s query contains any text from the chosen variable.
Is Not Present — Checks whether the user’s query doesn’t contain any text from the chosen variable.
Greater than — Use this when you want to check whether the user’s number input is greater than the number under the chosen context variable.
Less than — Use this when you want to check whether the user’s number input is less than the number under the chosen context variable.
Is True — Checks whether the chosen variable is true for the user’s query.
Is False — Checks whether the chosen variable is false for the user’s query.
Equals (Number) — Checks whether the user’s query contains the specified number from the chosen variable.
Equals (Number) — Checks whether the user’s query contains the specified number from the chosen variable.
Note: The Recognize Logic field on the pop up displays description about selected condition for this dialog.
9. Click on the Save button after adding the required conditions.
Note: To know more about creating a conversation using the available option nodes click here “Dialog Node Options”
10. Once you finish creating the first dialog node, you’ll be able to create additional nodes by clicking the ‘+’ button next to the created dialog node.
Note: To know more about creating a conversation using the various function nodes click here “Dialog Node Functions”
Note: To add more dialog nodes to the branch, finish setting up the condition inside the Recognise Dialog pop first. Setting up the first dialog consists of attributes like linking it to entities, intents, or variables.