Iciclecreek.Bot.Builder.Dialogs.Recognizers.QLucene

QLuceneRecognizer

This library provides a custom recognizers for .qna files using Lucene library.

Installation

To install into your project run the cli:

dotnet add package Iciclecreek.Bot.Builder.Dialogs.Recognizers.QLucene

In your startup code add:

ComponentRegistration.Add(new QLuceneComponentRegistration());

To add to your schema for usage in Bot Framework Composer from cli:

bf dialog:merge -p yourproj.proj -o your.schema

Library

This library adds Icicielcreek.QLuceneRecognizer which builds a lucene index from .qna json file and serves as an embedded recognizer for Bot Framework .qna files.

Sample Code

    var dialog = new AdaptiveDialog()
    {
        Recognizer = new QLuceneRecognizer()
        {
            KnowledgeBase = "foo.en-us.qna"
        }
        ...
    }

Sample Json

{
    "$kind": "Microsoft.AdaptiveDialog",
    "recognizer": {
        "$kind":"Iciclecrerek.QLuceneRecognizer",
        "knowledgeBase": "foo.en-us.qna"
    },
    "triggers":[
        {
            "$kind":"Microsoft.OnIntent",
            "intent":"QnAMatch",
            "actions":[
                {
                    "$kind":"Microsoft.SendActivity"
                    "activity":"${turn.recognized.entities.answer[0]}"
                }
            ]
        }
    ]
}

QLuBuild tool

The QLuBuild tool will process .qna files into the .qna.json files and automatically generate the supporting .dialog files bound to the output. It's useful, use it. :)