Custom question answering Json
Item Name: Custom Question Answering JSON
Category: Cognitive
Description: Allows handling the knowledge base (questions and answers) created in the Microsoft Azure portal in JSON format.
Functionality: Displays questions and answers from the knowledge base in JSON format, all in a single set. It is mainly used by developers to parse the delivered information and use it within the logic of the cognitive flow.
Requirements:
- Build Resource: Access the Microsoft Azure portal (portal.azure.com) in the Create Resource section. For more information, refer to Azure AD B2C.
- Build Items, each with their respective questions or possible answers: For further guidance.
- For some of the following fields, locate in the Azure AI / Language Studio / Custom Question Answering section under Deploy Knowledge Base. There, you should find the option to Get Prediction URL, which opens a pop-up window with the necessary data.
Configuration Fields (Required)
[Evaluation Expression]: A string field where an expression, either in the form of an entity or a quoted expression, must be entered for evaluation.
[QNA MAKER EndpointKey (Optional if AUTHORING_KEY is not available)]: A string field where authorization keys for connection configuration should be entered. Obtain this in the Microsoft Azure environment by locating the created resource and selecting the option Keys and Endpoint:
[OCP-APIM-SUBSCRIPTION-KEY]: A string field where the subscription keys for connection configuration should be entered. This is obtained from the pop-up window Get Prediction URL.
[PROJECTNAME]: A string expression field where the project name is defined. This is obtained from the pop-up window Get Prediction URL.
[DEPLOYMENTNAME]: A string expression field indicating the unique name that identifies the deployment instance. This is obtained from the pop-up window Get Prediction URL.
[Entity where the menu will be saved]: When the response is a menu, this drop-down list field allows selecting the entity where the menu will be saved.
[Intent to execute in case of an error. The action flow will be interrupted, and the error description will be passed to the selected intent with the name: 'QNA_MAKERV2_ERROR.]: A drop-down list where the entity name for error handling should be selected.
Custom Question Answering Implementation Example
--Prerequisites:
Created Resource: xxxxxx
Item: Menu
Response:
- Other available plans
- Go to Offers
--Input
[Evaluation Expression]: 'Hello'
[QNA MAKER EndpointKey (Optional if AUTHORING_KEY is not available)]: 'https://xxxxxx.cognitiveservices.azure.com'
[OCP-APIM-SUBSCRIPTION-KEY]: x56797vv98vc6cv789ooo
[PROJECTNAME]: 'Project_1'
[DEPLOYMENTNAME]: 'Production'
[Entity where the menu will be saved]: MENU_2
[Intent to execute in case of an error. The action flow will be interrupted, and the error description will be passed to the selected intent with the name: 'QNA_MAKERV2_ERROR.]: ERROR_Project_1
--Result: MENU_2 <--
{
"Answers": [
{
"Questions": [
"Double service package test",
"I want to hire a service package",
"hello, I need to hire",
"I want to hire multiple services"
],
"Answer": "Double:|Choose the most convenient offer [**here**](http://XXXXXX.cl/offers)|Choose an option or write your query",
"Confidence": 0.08220000000000001,
"QnaId": 16600,
"Source": "1000_Hire_26xxxx22",
"Metadata": {
"cat": "hj",
"sub": "gjhjhj",
"key": "rwer"
},
"Dialog": {
"IsContextOnly": false,
"Prompts": [
{
"DisplayOrder": 0,
"QnaId": 1234,
"DisplayText": "Other available plans"
},
{
"DisplayOrder": 0,
"QnaId": 1235,
"DisplayText": "Go to Offers"
}
]
},
"ShortAnswer": null
}
]
}