Related questions

Lex selects one intent for any one-word utterance that doesn't match a valid utterance

2018-10-11 19:37 Jay Gerring imported from Stackoverflow

For a few months, my team has been working on a Lex-based chatbot that's attached to a marketing website. The bot has a total of 32 intents that handle over 300 utterances right now. In an effort to reduce utterances that invoke error handling responses (and to make the bot more useful and friendly), we've implemented 3 intents in the following order over time:

1) a "Thank You" intent: responds with "you're welcome"-type messages when a user types some variant of "thank you".

2) a "No Answer Available" intent to handle known utterances (from our usability and internal user testing) that the bot isn't currently designed to answer, but that we want to refer the user to a human chat agent (e.g. "I don't have an answer for that, but type 'transfer' to talk to one of our human agents who can help."). This intent currently has the largest set of utterances in the bot.

3) a "No Response" intent that "eats" a small number of utterances from the user that don't really require a response and would probably go to error handling other wise and make the bot look dumb, e.g. "you're welcome", "ok, got it", "cool", "k" etc. IOW, it doesn't return any response for these to the user.

Until we implemented #3 (maybe coincidental?), if a user typed 1 or more nonsense words to the bot, it properly triggered error handling. After implementing #3, a single, alpha-only word of apparently any length that doesn't match a valid utterance is now triggering the "Thank you" intent.

Screenshot: bot says "you're welcome" to a nonsense word

Note that 2 nonsense words triggers the expected error handling messages. Also, single strings that contain symbols or numbers also trigger error handling as expected. Interestingly (and unexpectedly), text-based emoticons typed in the usual direction, e.g. :-) or :) or ;-) also trigger the "thank you" intent, but if you type them in the reverse direction, e.g. (-: they trigger error handling.

Additionally, 2 other intents handle a number of single-word utterances:

1) A "signoff" intent that handles words like: bye, goodbye, later, exit, quit, etc.

2) A "hello" intent that responds to: hi, hello, hey, what's up, what's happening, etc.

So the "Thank you" intent has neither the largest number of utterances in the bot nor the largest number of 1-word utterances (just "thanks", actually) both which I have seen in various forums as reasons why a Lex intent would attract incorrect utterances.

I've tried searching for answers on this phrasing the question any way I can think of and haven't found anything.

So, though it's not a product show-stopper, it's making me a little crazy, and I thought I'd post here to see if anyone has run into anything like this in their Amazon Lex journeys and had any ideas.

Thanks!

See also questions close to this topic

  • Run cron jobs in amazon web services for codeigniter function

    I want to call codeigniter function for every minute in amazon web services cron jobs.How do i do it.An help will be appreciated.

  • Configure ssl using aws CA

    I Used ACM to validate site for ssl.Instance is ec2 but domain provider is different and i have pointed domain correctly.

    Then i want to issue ssl for that instance or that pointed domain.so for that i have issued ssl using Aws using DNS and then added cname records to domain record set but the issue is still pending validation error.

    Then i come to know that we have to issue certificate using CA and use private certificate for that.now i have some doubts:

    1) what was wrong in my previous method and why its still pending verification 2) what is difference between private certificate using CA and validate aws certificate using DNS 3) if i try using Private certificate CA then it will issue me ssl for this domain and it will be secured that domain by https

    Any help will be appreciated.

  • Parse schema of a dynamic dataframe in AWS Glue

    I have a dynamic dataframe in AWS glue which I created using the below piece of code. val rawDynamicDataFrame = glueContext.getCatalogSource(database = rawDBName, tableName = rawTableName, redshiftTmpDir = "", transformationContext = "rawDynamicDataFrame").getDynamicFrame()

    In order to get the schema of the above dynamic dataframe, I used the below piece of code.

    val x = rawDynamicDataFrame.schema.Now x is of type com.amazonaws.services.glue.schema.Schema.How can I parse the schema object?

  • Build new logic adapter

    I have 3 different types of csv files in which there is different data, I want to write 3 logic adapter for each csv files, in which I have to write code using conditions like if statement has confidence <007 then it should show matching information , elseif it should show other response else different repsonse .

    So how I can built logic for it using chatterbot?

  • Whatsapp chatbot developing

    Today I have a theoretical question about a chatbot developing targeting Whatsapp or Viber.

    I like to make whatsapp/viber chatbot with PHP/JS/Python which asks people questions and retrieves data.

    I don't need to use NLP or AI, just need a simple chatbot. My preference is Whatsapp chatbot and not using frameworks (ex:twilio) of some other stuff (except libraries).

    Do you have some tutorials, recommendations or information about whatsapp chatbot developing, because I didn't find some proper info about it except Whatsapp Business API?

    Thanks a lot for your help.

  • Broadcasting from Facebook page

    I have a Facebook page and I was using Chatfuel as a bot script to send some broadcasts but now I got over 10k users and I must pay for it so I thought about making my own broadcasting script...

    When I strated to read FB developer docs I found thay it works and made it work but the problem is that till the app is verified I can't send a broadcast to all fb page fans ... Recording to developer.facebook.com If your Facebook app is in development mode, broadcast messages will only be sent to users that have been granted the Admin, Developer, or Tester roles for your app.

    So what I'm asking right now is if there is any why to get a verified access_token to my Facebook page to make this work or if I want to verify my app how much does it take or and If you have any suggestions or other ideas to make it work?

    Lastly, can I get the access_token that Chatfuel for example uses to acces my Facebook page?

    Thanks

  • Filling data in response cards with Amazon Lex

    I figured out how to send a Close response that would send details to the user and close the intent.But I can't figure out how to mange other types of comunication from the lambda to the dialog flow.

    A.How can I send custom values or information that would appear in the response message so the user could chose ?.

    Basiclly I think that I am asking for a custom slot configuaration through the lambda, based on the user choises.

    B.Specifically I am also trying to send (or configure) custom values in response card through the lambda. Those values would be selected in the lambda function based on the user choises so far during the dialog. How can I do that ?

  • Capturing missed utterances

    Does anybody know if it is at all possible to capture missed utterances? I don't see the missed ones being logged into CloudWatch. I know that you can view them in the Lex Console after 24 hours, but I'm trying to capture them with data attached to them.

    As of right now the console only gives you what the missed utterances is, how many times it was said, and when the last time it was said. I want the rest of the "Data" attached to these missed utterances; for example what customer said it.

    Does anybody know if this is possible with AWS or the SDK(.NET) currently with a lambda or something like that?

  • Does Amazon services support on-premise hosting?

    We intend to develop a enterprise Bot using amazon lex that will fetch response from a SQL server, and display result along with visual presentation. Does Lex support on premise deployment?
    Will there be any challenges in using Lex vs Google Dialogflow (formerly known as api.ai)?

    Please suggest.