I have a website which links to a chatbot built on IBM Watson Assistant. There are some hyperlinks on the website that I want to trigger specific nodes/ intents the watson dialog.
Example: User clicks on "Provide feedback" link, the watson chatbot launches and based on the link the "provide_feedback" intent is recognised (thus preventing the user from needing to specify the intent after clicking the link).
Has anyone tried this before?
I also came across this requirement and want to mention another alternative here:
Instead of sending an input text that matches the intent of your desired node, you can also pass
Intents to use when evaluating the user input.doc
and tell the assistant to match it with confidence of 1.0.
I think this is a clean method, because you don't need to deal with disambiguation of your input text.
Then you don't need to send input text at all and the intent actually does not even need example phrases :-)
For example if you want to trigger a node that has the intent #provide_feedback
you can call this python example code:
send_message_to_chatbot(text="", intent="provide_feedback")
def send_message_to_chatbot(text="", intent=""):
message = assistant.message(
assistant_id=ASSISTANT_ID,
session_id=SESSION_ID,
input=MessageInput(
text=text,
intents=[RuntimeIntent(intent=intent, confidence=1.0)]
)
).get_result()
return message
Prerequisite is of course that the node is in the root branch of your dialog so it can be triggered.
The Watson Assistant service basically is used via a REST API. That API is invoked from the "Try it" pane in the workspace editor, from your dedicated application or maybe from widgets embedded into a website. The message call is used to send user input to Watson Assistant and to receive a chatbot response.
What you can do is to call the message API from your app and pass a specific term as input message. That term would match an intent and hence trigger a specific dialog node. As an example, if you have an intent "provide_feedback" defined for the phrase "user pressed feedback button" and you pass in exactly that phrase as input message, then the intent "provide_feedback" will match.
Related
I'm developing a Google Action through DialogFlow and a webhook (that will run on a Nest Hub) that I want to act like this:
the user invokes the action "Hey Google, talk to ACTIONAME"
through the Default Welcome Intent ("hooked" to my web service) the Action replies to the user and open a website
app.intent('Default Welcome Intent', conv => {
conv.ask('Hi! I'm opening your site')
conv.ask(new HtmlResponse({
url: 'https://MY_IOT_SITE'
}))
})
now, the user could be "silent" for mins or hours, but I'd like to prevent Google Actions to close the ACTIONNAME and return to the clock, while until now the action closes after a couple of minutes
Is it somehow possible?
Thank you.
This is not possible. The platform intentionally places an upper bound on how long an action can run without any user input. This is done so that an action cannot occupy the device longer than expected and prevent future inputs from unintentionally getting routed to your action rather than the Google Assistant.
You can take a look at additional guidelines when developing your web app.
Since your question refers to an IoT-related website, you may want to take a look at the Smart Home reference, which provides an alternative way to let users control smart home devices with their voice or built-in graphical widgets.
I am making an app that predicts clothes the user is wearing. It uses the Visual Recognition tool, too, and for Conversation and VR to be communicating, I attach the intent 'suggestClothing' or 'clothResult' to the cloth items it found. I use an entity for Conversation to recognize the cloth items and respond accordingly.
The flow should be as follows:
User: how do I look?
-classifies clothes-
App to conversation: clothSuggest blackJacket
Conversation to user: "You picked the black jacket! Try out the green shirt with this outfit and show me how you look."
-classifies clothes-
App to conversation: clothResult blackJacket greenShirt
Conversation to user: "You look great in that outfit!"
All nodes have multiple responses as all clothes are in pairs. Either the user is wearing one or the other, and Conversation will then always suggest it's match.
Conversation flow looks like this
I also attempted this. Here sq123 is suggestClothing (first intent) and cq123 is clothResult:
This works fine in 'Try it out', too, but in the app, it immediately exits the branch on 'clothResult item1 item2' and matches with other conditions in the app.
What's the best way to optimize my flow to make it work in the app?
A typical reason why it works in "Try it out" and not in the app is that the context object is not returned properly.
When the app invokes the message method of the Watson Assistant API, a context object is passed. The calls are stateless and everything needed for Watson Assistant to continue a dialog is included in the context object. Thus, when your app retrieves the results from the message API, it needs to save the context and pass it back to Watson Assistant the next time the message method is invoked again (for that session and user).
I want to start Form conversation chat in Microsoft Bot framework.
if(user says hello)
{
reply = what u want to listen hi or hello
if(user says order)
{
reply= start a formbuilder.form with order form workflow
}
if(user says hello)
{
reply= hello
}
}
My problem is the first thing i do always works
example: if i say first chat line as order it starts order form but it never goes to the normal conversation mode even if the form ends.
if i start hi then it always goes in hi mode never goes or create order form on typing order.
Need it to be dynamic
you can use below code to end your conversation when you are in a dialog or a conversation flow ends.
context.Done<object>(new object());
or
context.Done(true);
do let me know if you need any help further
Per my understanding, you want start specific dialogs while triggering different words like "hollo" for greeting dialog, "order" for form dialog.
I think there are two methods to achieve this in C#:
You can leverage Recognize intents to implement LUIS, which can identify your users' intent from their spken or textual input, or utterances. Trigger specific dialogs for each LUIS intent.
For this solution you can refer to the official document Recognize intents and entities with LUIS using a prebuilt domain for details, and refer to https://github.com/Microsoft/BotBuilder-Samples/tree/master/CSharp/intelligence-LUIS for a sample for your reference.
You also can build Global message handlers using scorables in your bot application for yourself. With which, you can route users to certain fuctionality by using words like "help," "cancel," or "start over" in the middle of a conversation when the bot is expecting a different response.
Please refer to https://github.com/Microsoft/BotBuilder-Samples/tree/master/CSharp/core-GlobalMessageHandlers for the sample for this solution.
Hope it helps.
In the DialogFlow console, you can set an input and output context for an intent.
I would like to have the same functionality using the Actions SDK.
Since I can set the context in my fullfilment webhook in the code, the output SDK is covered.
However, how do I set the input context for an intent to only trigger if that context was fetched?
I could not find this in the documentation examples.
I don't believe you can in the same way, which is part of why they added Dialogflow.
Although as part of your response you can indicate the ExpectedInput for the next Intent, the documentation for the ExpectedIntent makes it clear that your Intents are only used for speech biasing - you'll always get the built-in TEXT Intent.
In fact, this documentation says that you'll only get custom Intents for the initial call - later Intents are always triggered with a built-in one.
Update: One thing you can do is to include information in the response that will be sent back to you in the next request.
If you're using the JavaScript SDK, this is done with the second parameter to ask() and is available via the getDialogState() function.
If you're sending back JSON, this is done using the conversationToken attribute and is available in conversation.conversationToken in the request object you're sent the next round.
I'm using Dialogflow (previously API.ai) to create an Actions on Google app.
Using Dialogflow I have set up a custom Entity to highlight single words from multiple Intents.
ie. mashable, recode, bbc sport are all words picked up as Entities.
The fulfilment sends a post webhook to an api I created.
When using Dialogflow it sends off
"parameters": {
"news-agent": "BBC Sport"
},
Which is fine, I set up API to detect the parameter and when using Actions on Google simulator
"parameters": {
"news-agent": ""
},
The parameter is blank, I don't see anything in the documentation about why this is happening.
Could someone help?
It's possible that Actions on Google is not picking up a value for the "news-agent" slot.
Try making the "news-agent" slot required in Dialogflow and define a re-prompt question for it.
That way, the event won't be sent to your app unless the "news-agent" slot is filled.