trying to make an Intent for my first Conversational Assistant with a couple of parameters. Everything is build in Dialogflow and for some of my functionality I use fulfillment.
The easiest way to describe my assistant is as a cinema tickets booking.
I need to store #number_of_tickets, #cinema, #movie. The last two (#cinema and #movie) are very easy to solve with entities, however struggle a lot with #number_of_tickets. All of this parameters should be mandatory.
My goal is to allow such kind of functionality:
Book me a ticket for Star Wars in IMax (ticket = 1 = #number_of_tickets, Star Wars = #movie, IMax = #cinema)
Book me 2 tickets for Star Wars in IMax(2 tickets = 2 = #number_of_tickets, Star Wars = #movie, IMax = #cinema)
Book me tickets for Star Wars in IMax -> How many tickets you want to book?
I tried with composite Entities but it doesn't work as I'm expecting.
How I can enable such kind of functionality in Dialogflow?
Is there a way to require #number_of_tickets to be between 1 and 5?
Above question can be achieved by following below steps:
Solution:
Step 1:
Create Entities for movie and cinema:
Cinema:
Movie:
Step 2:
Create an intent with three parameters:
(If you want alow only from 1 to 5 then you can create an entity like cinema and movies with entries from 1 to 5 and map the number of tickets parameter with the same)
Step 3:
Make all the parameter Mandatory and define prompts for all as below:
Movie Prompts:
Cinema Prompts:
Number of tickets prompts:
Result:
It may not be obvious, but the answer to both of your questions can be addressed by remembering that Intents capture what the user says, and not what you do with what they say. You can have multiple Intents to capture different phrases, and handle them in mostly the same way in your fulfillment. (And these do require that you have some kind of fulfillment webhook behind the scenes.)
How can I assume a default of 1 ticket?
There are a couple of approaches.
The first is that you can setup an Intent to capture phrases that suggest a single ticket and a different Intent that explicitly requires a number. So perhaps one Intent (possibly book.single) with
I'd like to get a ticket for Star Wars in IMax
How about the IMax showing for Star Wars
Let me see the Star Wars for IMax
and so forth. While you'd also have a different Intent (possibly book.multiple) with
I'd like 2 tickets for Star Wars in IMax
2 tickets please for the IMax Star Wars
etc.
In your fulfillment, you might have some function that takes three parameters
bookTickets( num, movie, cinema );
In your Intent handler for book.single, you'd call this with
bookTickets( 1, movieParameter, cinemaParameter );
while in the Intent handler for book.multiple, this would be done with
bookTickets( countParameter, movieParameter, cinemaParameter );
In each case, the fulfillment still ends up calling the same function to do the actual work, but the handler sets the values as appropriate.
How can I limit how many tickets get ordered?
This goes to something that you didn't address - what do you want to happen if they try to order more? Do you want to direct them to a group order conversation? Reject it outright? Whatever you want - make sure you've planned out this path of a conversation before you try to implement it.
In this case, this isn't something you want to enforce in the Dialogflow Intent. Handling it in your fulfillment is easiest. There, you would apply any logic for the various parameters and return a message, prompting them about their next reply.
Logic is best handled in the fulfillment.
There are actually a couple of places this is important in your example. In the first, how would you handle more than 5 tickets requested? In your handler for book.multiple, you would check how many are requested. If between 1 and 5 inclusive, you'd call the book function above. But if not, you might want to just send back a replying saying "I'm sorry, you can only reserve 5 tickets maximum. How many tickets do you want?"
You would also need a new Intent to capture the reply. It could be as simple as just taking a number, but to make sure it only gets applied when you send back this prompt, you may want to set an Output Context indicating you're prompting for the number of tickets and then set this as an Input Context for the Intent that collects a number.
Are there other cases where you might want to apply logic before booking tickets? Plenty - and in each case you will probably want to make sure there is a new Intent that can capture what they want. For example, what if the movie they want to book isn't showing in that cinema? What if you already know that movie is sold out?
But what about Followup Intents?
Followup Intents are generally best avoided in general for exactly the same reason - Intents represent what the user says, and shouldn't try to model your code logic. There is nothing special about a Followup Intent - you can create an Intent at the top level that works the same way, but will be valid only if you prompt for certain things. How? Contexts again!
So if you prompt with possible showtimes ("Do you want the 2:00, 4:00, or 6:00 show?") you could also include a context (say "promptTime", for example) and then have an Intent that expects this as an Input Context and matches the times.
Conclusion
Intents represent what the user says
Your fulfillment code does something with what they've said
You can prompt them for more or different information in your fulfillment
Contexts can shape what replies you expect
Multiple Intent Handlers can call the same functions to do the work
Related
Good day,
I am currently trying to create an intent that is able to put out responses depending on the user input. (The chatbot should be implemented on a website later)
Let's say we have an entity called cars with three entries: "Volkswagen" "Audi" "Ford".
Now when the user types in something with e.g. Audi in it, the response will correspond to this. Something like this: If Audi then give this response, if Ford then this response.
I couldn't find anything helpful yet.
Thank you in advance!
Remember that Intents represent what the user says and not how you handle this or how you reply. Although Dialogflow does offer the ability to respond to Intents, these aren't based on specific values that may appear in parameters.
There are, however, a variety of ways you can handle what you're doing, based on the rest of what you're trying to do.
Multiple Intents
One solution is to create an Intent for each type of thing that the user may talk about. You could then put the response you want for each in the response section of that Intent.
This is probably a bad approach, but may be useful in some ways. It requires you to duplicate phrases between the different Intents, which leads to a lot of duplication. On the upside, it does let you vary the replies, and truly represents the Intent of what the user is trying to say.
Using Parameters with Fulfillment
A better approach is to have a single Intent with many phrases representing what the users can ask. These phrases would have parameters of your Entity Type.
You can then enable Fulfillment for this Intent and write a Fulfillment webhook for the Intent that would look a the value of the parameter and send back an appropriate reply.
Using Parameters with DetectIntent
Since your ultimate goal is to embed this on a website, it may be more appropriate to have your website show something different based on what the user has said. (For example to show a picture of the car in another pane or links to different pages, to use your example of cars.)
In this case, your chat client (or a proxy) would be calling the DetectIntent API. You can structure your Intent similar to above, with parameters of the Entity Type, and the reply that is sent to your client would contain the Intent along with the value of the parameter. Your client then can check the value of the parameter and change the display accordingly.
My agent’s goal is to ask 6 questions to make a price estimation of 2 different products.
It asks the 1st question: ‘do you want to make a price estimation for product1 or product2 ?’
And then, it asks 5 questions about the product’s characteristics (i.e. the product’s age, size…) to finally display the price estimation.
All flow made on Dialogflow UI with an inline editor for the fulfillment to calculate and then save data on Firebase DB.
My question -> Is it better to create :
1/ Create only 2 intents : product1 and product2 ? Then in those intents, ask the others 5 questions in ‘action and parameters’ section by filling parameters, entities and values.
2/ Create 6 differents intents for the 6 questions, with parameter value for each ones.
On my side :
I have tested the 1/, I’m able to retrieve all data on my Firebase DB but my 2 products have similar entities and parameters values (‘size’ for example) and my agent make errors’ flow to the wrong intent.
I have tested the 2/ (I think it’s the better way because all intents are more efficient) I’m able to make the good flow. But at the last question, I don’t know how to retrieve previous parameters values to make the price estimation.
Thanks a lot for the help ;)
Remember that Intents are designed to capture what the user is saying - not necessarily the logic behind how you intend to handle that. While you can use things like Contexts to keep track of state, it is sometimes better to handle that with your logic. (See Thinking for Voice: Design Conversations not Logic)
You should be storing anything you wish to remember between Intents in the parameters of a long-lived Context.
I have created a pizza bot in dialogflow. The scenario is like..
Bot says: Hi What do you want.
User says : I want pizza.
If the user says I want watermelon or I love pizza then dialogflow should respond with error message and ask the same question again. After getting a valid response from the user the bot should prompt the second like
Bot says: What kind of pizza do you want.
User says: I want mushroom(any) pizza.
If the user gives some garbage data like I want icecream or I want good pizza then again bot has to respond with an error and should ask the same question. I have trained the bot with the intents but the problem is validating the user input.
How can I make it possible in dialogflow?
A glimpse of training data & output
If you have already created different training phrases, then invalid phrases will typically trigger the Fallback Intent. If you're just using #sys.any as a parameter type, then it will fill it with anything, so you should define more narrow Entity Types.
In the example Intent you provided, you have a number of training phrases, but Dialogflow uses these training phrases as guidance, not as absolute strings that must be matched. From what you've trained it, it appears that phrases such as "I want .+ pizza" should be matched, so the NLU model might read it that way.
To narrow exactly what you're looking for, you might wish to create an Entity Type to handle pizza flavors. This will help narrow how the NLU model will interpret what the user will say. It also makes it easier for you to understand what type of pizza they're asking for, since you can examine just the parameters, and not have to parse the entire string again.
How you handle this in the Fallback Intent depends on how the rest of your system works. The most straightforward is to use your Fulfillment webhook to determine what state of your questioning you're in and either repeat the question or provide additional guidance.
Remember, also, that the conversation could go something like this:
Bot says: Hi What do you want.
User says : I want a mushroom pizza.
They've skipped over one of your questions (which wasn't necessary in this case). This is normal for a conversational UI, so you need to be prepared for it.
The type of pizzas (eg mushroom, chicken etc) should be a custom entity.
Then at your intent you should define the training phrases as you have but make sure that the entity is marked and that you also add a template for the user's response:
There are 3 main things you need to note here:
The entities are marked
A template is used. To create a template click on the quote symbol in the training phrases as the image below shows. Make sure that again your entity is used here
Make your pizza type a required parameter. That way it won't advance to the next question unless a valid answer is provided.
One final advice is to put some more effort in designing the interaction and the responses. Greeting your users with "what do you want" isn't the best experience. Also, with your approach you're trying to force them into one specific path but this is not how a conversational app should be. You can find more about this here.
A better experience would be to greet the users, explain what they can do with your app and let them know about their options. Example:
- Hi, welcome to the Pizza App! I'm here to help you find the perfect pizza for you [note: here you need to add any other actions your bot can perform, like track an order for instance]! Our most popular pizzas are mushroom, chicken and margarita? Do you know what you want already or do you need help?
Suppose i want to make a pizza ordering DialogFlow agent. To order a pizza we need 3 things: size, type and toppings.
If we want to go with follow-up intents approach rather than using entities then there will be so many combinations in which user might provide the information.
1: i want a pizza -> no info
2: i want small pizza -> size
3: i want small cheese pizza -> size and type
4: i want small cheese pizza with olives -> size, type and toppings
5: i want small pizza with olives -> size and toppings
...
and so on
How to solve this problem?
There will be so many combinations if we have more entities (2^n combinations)
Note 1: cannot take entities and slotfilling option as there are so many problems if we go down that road, like re-prompts loop, validation etc.
Is there any better solution?
Note 2: If we use entities, mark them required, and set prompts then many times if it does not get desired input from user it get stuck in re-prompt loop, i.e it keeps asking user same (or random) prompt for same entity. In my use case, it is bad for user experience. If we use follow-up intents instead, then we can set fallback intents for all those intents which solved this problem. (please note that this is just example of the use case)
This is another example of why I used follow-up intents, it solved my date capturing problem as well. I took #sys.date.recent and set a fallback intent to capture inputs like last week, last month etc, this was not possible using slots.
First, remember that Intents should reflect what the user says, not necessarily what you are doing.
On the surface, it isn't clear why slot filling (either with fulfillment or using built-in prompts) won't meet your needs. Since you've indicated that all three bits of information are necessary (size, type, and toppings), you can mark them as such in your phrases and Dialogflow will prompt for the missing information until it gets everything.
You almost certainly do not want to use Followup Intents. These are good when you always have a specific response that you send that will always have a very narrow set of replies from the user, but are very poor if the response from your action will prompt the user to reply in many different ways.
Instead, I would use a related concept: Contexts. (At least if you're not going to use slot filling.) When you ask the question of what they want, set a Context so you know they are. Then have one or more Intents that have this as an Input Context that accept the various things the user might say. Your webhook should see if you have the information you need and, if not, prompt them what else you're looking for. At the end, prompt for a confirmation, but they may say something that adjusts the order.
I know that i can make a none intent to cover some of these, however we cannot just create every nonsense question a person could ask.
Or even if someone types in a 50 word statement. The bigger problem is that if we get a query to LUIS, it is assigning it an intent that is not correct, without even having identified any entities either.
What to do?
To handle these cases, it would be better to add more labeled utterances to your other intents and occasionally add the stray utterances to the None intent. When the model is better for predicting your non-None intents, the better predicting of None intents also accompany this (LUIS attempts to match to an intent rather than cutting intents out).
If intents are triggering without any entities being recognized (and thus you believe the wrong intent has been triggered), this should be handled at an application level, where you would then disambiguate the intents back to your users. If you've set the verbose flag to true, then you could take the top three scoring intents and present those back as options to your user. Then you can move back into the proper dialog.
After you've moved into the intent/dialog they meant to access, you can conduct a programmatic API call to add that utterance to the intent. Individually adding labeled utterances can be problematic (the programmatic API key has a limit of 100,000 transactions per month, and a rate of 10 transactions per second), so you can instead aggregate the utterances and conduct batch labeling. An additional bit of info; there is a limit of 100 labeled utterance per batch upload.
Adding to the Steven's answer - in the intent window, you have the Suggested Utternaces tab - this is also a hint for the algorithm, kind of reinforced learning approach.