Call Knowledge Collection Using service externally

Hi Team,

I would like to know that how can i call the knowledge collection externally using the service.

Do you provide any Api in which it find the intent and the response of the query that we call.

Please response as soon as possible

Hi @renisweta1094,

For knowledge collection, there is no external API that is exposed to retrieve the response for the FAQ.

However, you may integrate the WebHook channel into the bot which would retrieve the required information as per the request provided that works within the bot.

To integrate the webHook channel, please refer the following link:
https://developer.kore.ai/docs/bots/bot-builder/adding-channels-to-your-bot/adding-webhook-channel/

Regards,
Yoga Ramya.

@renisweta1094, The platform does provide a find intent API which returns the intent identified based on the input provided. This API provides the information on intents identified by the engines FM, ML and KG depending on the configuration at the bot and also the final winner announced by the final resolver engine.

Please refer to the below link for details on findIntent public API:
https://developer.kore.ai/docs/bots/apis/find-intent-api/

Hi @Subrahmanyam ubrahmanyam,

I have checked the intent detection api that provided by kore.ai
But when ever I type a query regarding the knowledge collection i did not get the intent(question ) that is detected .

Need your help in that case. i am sharing the output of request that getting when i call the intent detected api with a question in knowledge collection.

{
“request”: {
“input”: “unable to modify questionnaries”,
“streamName”: “DbFaqBot”
},
“response”: {
“usedtime”: 344,
“debugTitle”: “Intent Match Successful: ‘’”,
“result”: “successintent”,
“messageStoreId”: “xxx”,
“bot”: “DbFaqBot”,
“botid”: “st-40cf2ad3-ffed-535f-9b18-e62795297adb”,
“identifiedVia”: “none”,
“input”: [
“unable to modify questionnaries”
],
“output”: "I’m Sorry, this task is not accessible to you. Please contact Administrator to get access. ",
“isDeveloper”: true,
“language”: “en”,
“userId”: “u-69dbebf5-f448-565f-99ba-84ab654eb135”,
“time”: “2019-03-19T05:10:50.288Z”,
“_id”: “f-4d23d6bd-9f18-541f-b1bc-6c871d01f0c3”,
“toneAnalysis”: {
“dialogTone”: [
{
“tone_name”: “angry”,
“level”: 1,
“count”: 2
},
{
“tone_name”: “positive”,
“level”: 1,
“count”: 4
}
]
},
“ml”: {
“possible”: [
{
“task”: “FaqCategory”,
“state”: “published”,
“score”: 0.4430567681026193,
“matchType”: “possible”
}
],
“eliminated”: [
{
“task”: “Welcome”,
“state”: “published”,
“score”: -1.5034094250492283,
“matchType”: “unLikely”
},
{
“task”: “Farewell”,
“state”: “published”,
“score”: -2.0583445604186084,
“matchType”: “unLikely”
}
],
“namedEntityRecognition”: []
},
“fm”: {
“eliminated”: [
{
“mlScore”: -2.0583445604186084,
“count”: 0,
“score”: 0,
“activityType”: 1,
“exactcount”: 0,
“labelsize”: 1,
“ignorewords”: 0,
“mask”: 0,
“words”: {
“farewell”: []
},
“priority”: 0,
“tense”: 0,
“mainRoles”: 0,
“foundFmEngine”: true,
“matchType”: “eliminated”,
“task”: “Farewell”,
“state”: “published”,
“foundVia”: “wordMatch”
},
{
“mlScore”: 0.4430567681026193,
“mlSentence”: 0,
“count”: 0,
“score”: 0,
“activityType”: 1,
“exactcount”: 0,
“labelsize”: 1,
“ignorewords”: 0,
“mask”: 0,
“priority”: 0,
“tense”: 0,
“mainRoles”: 0,
“scoreBreakdown”: {
“wordMatch”: 0,
“exactWords”: 0,
“coverage”: 0,
“sentenceBonus”: 0,
“positionBonus”: 0,
“orderBonus”: 0,
“spreadBonus”: 0,
“roleBonus”: 0,
“faqQuestionBonus”: 0,
“mlMatchBonus”: 0,
“matchBonus”: 2000,
“phraseJoinPenalty”: 0
},
“foundFmEngine”: true,
“words”: {
“faqcategory”: []
},
“matchType”: “eliminated”,
“task”: “FaqCategory”,
“state”: “published”,
“foundVia”: “wordMatch”
},
{
“mlScore”: -1.5034094250492283,
“count”: 0,
“score”: 0,
“activityType”: 1,
“exactcount”: 0,
“labelsize”: 1,
“ignorewords”: 0,
“mask”: 0,
“words”: {
“welcome”: []
},
“priority”: 0,
“tense”: 0,
“mainRoles”: 0,
“foundFmEngine”: true,
“matchType”: “eliminated”,
“task”: “Welcome”,
“state”: “published”,
“foundVia”: “wordMatch”
},
{
“count”: 0,
“score”: 0,
“activityType”: 1,
“exactcount”: 0,
“labelsize”: 1,
“ignorewords”: 0,
“mask”: 0,
“words”: {
“callingintentapi”: []
},
“priority”: 0,
“tense”: 0,
“mainRoles”: 0,
“foundFmEngine”: true,
“matchType”: “eliminated”,
“task”: “CallingIntentApi”,
“state”: “configured”,
“foundVia”: “wordMatch”
}
]
}
},
“streamId”: “st-40cf2ad3-ffed-535f-9b18-e62795297adb”,
“streamName”: “DbFaqBot”,
“seqLogId”: “f-4d23d6bd-9f18-541f-b1bc-6c871d01f0c3”,
“_id”: “f-4d23d6bd-9f18-541f-b1bc-6c871d01f0c3”,
“name”: “DbFaqBot”,
“input”: [
“unable to modify questionnaries”
]
}

Please reply on this as soon as possible.

Thanks.

@renisweta1094
Could you please pass the command “cheat developer off” in the talk to bot on bot builder, then pass the same utterance and let us know if the FAQ is identified?

Please make sure the knowledge collection is published and made available to the end users .

Hi @Subrahmanyam
I am getting below error while trying to hot the find intent API from postman.
{
“errors”: [
{
“msg”: “Permission denied. Scope is incorrect!”,
“code”: 4002
}
]
}

What permission is required here ? and it will be required on app level or bot level ?

@mohini.chauhan

As mentioned in the documentation, bot level scope is required for findintent api with ‘Intent and Entity Detection’ permission.

https://developer.kore.ai/docs/bots/api-guide/find-intent-api/