I wonder if in future the prebuilt entity personName will be available in italian language. I notice that this entity works very well also for Italian name inside an english sentence. I tested this in a LUIS Project in english language.

To create Dispatch LUIS we need a GUI in LUIS.ai itself so that it would be more user friendly than the CLI command "npm i -g botdispatch". Also, if there is any errors we can easily understand with friendly messages. I faced an issue recently while executing the above command and raised it in Microsoft community. I used latest nodejs (v12.13.1) and npm (v6.12.1) on Win10 Enterprise OS.

https://social.msdn.microsoft.com/Forums/en-US/843a38db-85b1-4a7e-a08a-7d2687dff170/is-there-dispatch-web-tool-for-integrating-luis-and-qna?forum=AzureCognitiveService

It's not possible to delete all utterances in 'Review endpoint utterances'. You can only do this one by one. This is not working for me, because I've >5000 utterances in 'Review endpoint utterances'.

So add the possibility to remove all utterances under Review endpoint utterances in once.

Choose a pattern that involves a {personName}, e.g. one from the calendar app;
^does {personName} have time at {Calendar.StartTime}

If you enter the utterance of "does george smith have time at 12:00am" then everything works as expected. However, if you change the last name to be a city, then it fails to match. E.g. "does george washington have time at 12:00am" then it fails to find the name (and can fail to match the pattern).

apps created using preview.luis.ai will have schema version 6, when exporting the app as JSON in order to generate the C# class, LUISGen cannot generate C# code anymore.

luis.ai page login screen should accepting login id from me. Currently it uses my office login id automatically and logs-in. I could not use my personal login id.

Getting error during Migration:
BadArgument: Failed retrieving user's Azure information, retry again later.
Please help - I tried this a week ago - how much later is "later"?

without a human support for interpreting and analysing a hided or coded sentence or phrase ,you will never reach an acurate and precise interpretation for a compounded speach on live or recorded .

As of now i can send single sentence as input to Predict API. There should be a way to send a batch of sentences and get corresponding intent of each sentence.

Because a lot of words used in utterances and entities are split in German, a new tokenizer is available for this language. In Dutch the problem is exactly the same. It would be really helpful if the other tokenizer was also available in Dutch.

If I want to use Special Characters like ö,ä,ü and ß in the text of my Utterances and the Applications Culture Settings on de-de I will get the Error-Message:
{
error: {

code: BadArgument,

message: Failed to parse application json. Parameter name: applicationJsonString

}
}

It just works sometimes when I use the Special Characters in the last few Utterance-Text.

Out of scope recognition: A recent change of LUIS backend impacted recognition of out of scope and long utterances very badly.
Steps to reproduce:
take your LUIS model and test with any text paragraph taken from another domain, web site or what ever.
Unexpected result is, that many intents are hit with a very high score far above 90%. This behaviour is not acceptable and had been better in the past.

Phrase list cannot be added to sub-component of an entity, for example in the LUIS tutorial to add the descriptor for Size entity that is part of Order entity, the following instruction does not work.
to add a descriptor, select the + in the Descriptors for Size area, then select Create new phrase list.

I am using pattern recognition to catch entities with a variable size. Here are situation that i am trying to catch

1- {entity1} (has| had| have) [the] {entity2}

2.1- {entity1} (has| had| have) the {entity2}

2.2- {entity1} (has| had| have) {entity2}

i tried the 1 pattern or the 2.1 and 2.2 at the same time.
The problem is that when i enter: "Person have the properties"

the entity2 is marked as "the properties" instead of just "properties"
It would be nice to have a way to set priority in the patterns to avoid this problem

Consider the following scenario. I have two intents:

1) add-task "add task [Task]"
2) assign-task "assign it to [Person]".

I would like to enable the utterance "add task clean break room and assign it to Jim"

Of course this can be done today either directly in the intent or through entities but not in a dynamic fashion. What if I want an additional intent:
3) complete-task "complete it"

This would require n^2 permutations for every nestable intent.

On clicking any directories listed in the directories for my user account, nothing happens. I can see a 400 error when viewing in the web debug console