We have taken you by the hand thus far and walked you through the code you were asked to produce step-by-step. We will still be providing useful information for you to learn more about developing conversational agents, but now we will change gears a bit and leave more for you to figure out yourself, too. Remember that you can find useful information and links on the Project Background Knowledge page.

Dialogflow

Conversational patterns set expectations about what actors participating in a conversation will do, but users often will not quite meet these expectations and make moves that do not fit into the active pattern. The conversational agent will need to be able to handle such “unexpected” moves from users. “Unexpected” here means that these moves do not fit into the currently active conversational pattern, not that such moves should not be expected from users. Two intents that we should actually expect to match with what a user could say are an appreciation intent and an intent for checking what the agent is capable of. You should add the following intents to your Dialogflow agent, making sure that you use the intent labels specified below:

1. Add an appreciation intent to match user expressions of appreciation or gratitude. The intent should, for example, match with a “thank you” phrase.

2. Add a checkCapability intent that enables a user to inquire about what your agent can do. The intent should, for example, match with a phrase such as “What can you do?”.

Fallback intent

Another “unexpected” intent does not have its origin with the user, but rather is a result of misunderstandings that arise due to speech recognition problems. If, in our case, Dialogflow is not able to transcribe what a user says and classify it as a known intent (one of the intents created for your Dialogflow agent), it will classify what the user says as a default fallback intent. Or, in other words, the default fallback intent is matched when your agent does not recognize an end-user expression. Check out https://cloud.google.com/dialogflow/es/docs/intents-default#fallback. You do not need to add a fallback intent, as it is already available when you create your Dialogflow agent.

The default fallback intent is a special intent for several reasons. It is matched if Dialogflow cannot match user input with any of the other intents known by the agent. But Dialogflow will also match with this intent if user input matches with training phrases that you provide for the fallback intent. You can add training phrases to the fallback intent that act as negative examples to make sure these phrases are not matched with any other intent. There may be cases where end-user expressions have a slight resemblance to your training phrases, but you do not want these expressions to match any normal intents.

Because phrases that are completely unrelated to the topic of our recipe recommendation agent will be classified as a fallback intent when they do not even vaguely resemble any of the training phrases that you add to the intents you create, it is not necessary to add such phrases as negative examples. Instead, it is more useful to think of phrases that a user might say and are similar to some of the training phrases used for your Dialogflow agent that should not be matched with any of your agent’s intents. It is, however, not that easy to come up with such phrases as the cooking domain is very extensive, and it is not completely clear what a recipe recommendation should be able to understand and/or handle. Perhaps it is best to include, as a design decision, some phrases that the agent clearly will not be able to handle (e.g., because of limitations of its database). An example that comes to mind is a user saying something like “I don't want anything that has a lot of calories”. It will not be easy to handle such a request because the information that is available about calories related to the recipe in the database is very limited at best. However, arguably, even if the agent is not able to handle the request very well it should at least be able to understand the request (the Dialogflow agent should be able to make sense of it) and provide an appropriate response such as an apology, for example, that it is unable to process a request like this. Perhaps a better example would be a user saying “I am interested in rabbits” in the sense of “I have an interest in rabbits”. These statements are somewhat similar to expressing a preference about a recipe (“I’d like a recipe with rabbit”) but their meaning is quite unrelated to requests about recipes. In any case, it’s up to you to make your mind up about what kind of training phrases should be used for the fallback intent.

Throughout the project, you should keep checking the validation page for issues and update the fallback intent by adding negative examples when they come to mind (e.g., when you add more training phrases for other intents).

Prolog and Patterns

Repair

When a conversational agent does not understand a user, it needs to have the conversational competence to deal with the situation. Here, we will make two important distinctions related to the type of misunderstanding. The first case is that the agent is not able to match what a user says with an intent and the Dialogflow agent matches with a fallback intent. The second case is quite different. Here the Dialogflow agent is able to make sense of what a user says and matches it with one of the intents created for that agent. The conversational agent, however, has trouble handling the intent within the conversational context as defined by the active conversational pattern. In other words, the intent does not fit into that pattern and the question is how the agent should deal with this. We provide two repair mechanisms using (somewhat special) patterns again for the agent to deal with each of these situations.

Responding to a fallback with a paraphrase request

An example of a user expression that will not be recognized by our Dialogflow agent is the following:

U: Have you read the Hobbit?

A: What do you mean?

In this case, the user expression clearly does not match with any of the intents that we created for our Dialogflow agent and the agent will not understand it (a fallback intent will be received from the Dialogflow agent). In Moore and Arar, 2019’s taxonomy, this is a b12 pattern. In this pattern, the agent responds to a misunderstanding by asking the user to rephrase, i.e. it makes a paraphraseRequest move. Retrieve the intent label that you should use for the first user move by inspecting the Action section of the default fallback intent in your Dialogflow agent (and make sure this label is of type atom).

• Add a b12 pattern to the patterns.pl file.

Don’t forget to add textual responses in the responses.pl for the paraphraseRequest intent. You can use the Responses section of the fallback intent in your Dialogflow agent for inspiration.

Responding to an out of context user intent

Now suppose that the user said something that Dialogflow can match with one of the agent’s intents but that intent does not fit into the active conversational pattern. An example for handling a situation like that is the following:

A: What recipe would you like to cook?

U: Hey there.

A: I am not sure what that means in this context.

We call the second move of the user an out of context move. The agent could expect many intents to fit here but a greeting such as the user expression in the example is out of context and does not fit. In principle, there can be many intents that simply do not fit the context of the conversation. The agent should respond to such out of context intents with a response that indicates there is a mismatch with the context (at least from the agent’s perspective). We call the last agent move therefore a contextMismatch. The pattern that we want to implement here as a repair mechanism consists of the out of context user intent as first move and the contextMismatch as second move. The identifier we use for this pattern is b13. This pattern is somewhat different from other patterns as we only know for the first [Actor, Intent] pair that the Actor must be instantiated with user but there is no specific intent we can use to instantiate the Intent parameter. For this reason, as it should be possible to instantiate this paramater with in principle any intent, we should keep this variable and we will not instantiate it at all. To complicate things even further, to enable the agent to shape its response specifically based on the Intent that is out of context, we want to pass this parameter on to our contextMismatch move. Providing a generic type of response as in the example is very dissatisfying given that the conversational agent has so much more information it can use to shape its response. To make this possible, we will simply add the parameter to the intent label and use contextMismatch(Intent) for the second agent move.

• Add a b13 pattern to the patterns.pl file using Intent and contextMismatch(Intent) as the intent labels for the first and second dialog move.

Make a contextMismatch response in responses.pl.

paraphrase requests are determined depending on the context of the conversation (the pattern that is currently active). Under the paraphrase request section in responses.pl, you should create ‘text/3’, which specifies the active pattern, the agent response name, and what the agent should say. Fill all these in. 

% Intent: paraphrase request

text(c10, paraphraseRequest, ""). % we don't care exactly what user said. 
                                        we got some response.

text(a50recipeSelect, paraphraseRequest, " ") :- % this should be used 
                                if the number of filtered recipes is greater than 0
                                
text(a50recipeSelect, paraphraseRequest, " ") :- % this should be used 
                                if the number of filtered recipes is  0                                
	
text(a50recipeConfirm, paraphraseRequest, "").

text(c43, paraphraseRequest, "").

Appreciation 

A simple example of a pattern where a user first expresses its appreciation and the agent receives this well is the following:

U: Thanks

A: You're welcome.

In Moore and Arar’s taxonomy, this classifies as a b42 sequence closer appreciation pattern. Implement this pattern in patterns.pl. You should use the intent labels appreciation and appreciationReceipt. Add phrases the agent can use for expressing the receipt of the user’s appreciation in the responses.pl file.

Checking capabilities

When a user wants to know what your agent can do for it, i.e., check what capabilities it has, the agent should be able to provide an appropriate reply. The key challenge here is to fill in the ___ in the example below. What would be a good response to such a general request for information of a user? The capability check should give a user enough guidance to understand how to talk to the agent or, even better, ideally also to ask more specific questions about its capabilities, for example, “tell me more about the recipe features you know about” (cf. Moore and Arar, 2019).

Example:

U: What can you do?

A: I can ____.

In Moore and Arar’s taxonomy, this classifies as a c30 pattern for a general capability check. Implement this pattern in the patterns.pl file. You should use the intent labels checkCapability and describeCapability. Define the agent’s response in the responses.pl file.

Visuals

Nothing we ask you to do here for this capability. It’s up to you.

Test it Out

Say random stuff like “the sky is blue” in all steps and see how your agent responds and if it freezes or if you are allowed to continue.

Before the repairs, the agent froze if it did not understand and could not proceed. Now you should be able to say something random, it says it does not understand and you could then continue the pattern.