wechaty-qnamaker
Version:
QnAMaker.ai Plugin for Wechaty
179 lines (118 loc) • 7.23 kB
Markdown
# WECHATY-QNAMAKER
[](https://www.npmjs.com/package/wechaty-qnamaker)
[](https://github.com/wechaty/wechaty-qnamaker/actions?query=workflow%3ANPM)
[QnA Maker](https://qnamaker.ai) is a cloud-based Natural Language Processing (NLP) service that easily creates a natural conversational layer over your data.
Wechaty QnAMaker can be used to find the most appropriate answer for WeChat users, from your custom knowledge base (KB) of information.
[](https://github.com/wechaty/wechaty-qnamaker)
[](https://github.com/Wechaty/wechaty)
[](https://www.typescriptlang.org/)
## INTRODUCTION
Wechaty QnAMaker Plugin helps you to answer questions in WeChat with the power of <https://QnAMaker.ai>.
## REQUIREMENTS
1. Node.js v12+
1. Wechaty v0.40+
1. This QnAMaker Plugin
1. Azure Subscription for Cognitive Service
1. QnAMaker Knowledge Base (KB)
## USAGE
To use the plugin:
```ts
import { WechatyQnAMaker } from 'wechaty-qnamaker'
const config = {
mention: true, // default true: require mention the bot in room.
room: true,
contact: true, // enable direct message.
/**
* Language of Questions & Score of Answers
*/
language: 'english',
scoreThreshold: 50, // minimum score for the answer
/**
* QnAMaker Service API
*/
endpointKey: '705a3468-12bb-4e10-a314-7daa947f18d6',
knowledgeBaseId: '254e33ad-ca6d-405d-980d-dbd3615e2605',
resourceName: 'wechaty',
}
const QnAMakerPlugin = WechatyQnAMaker(config)
const wechaty = new Wechaty()
wechaty.use(QnAMakerPlugin)
```
We also provide a Vorpal Command:
```ts
import {
WechatyVorpal,
WechatyVorpalConfig,
} from 'wechaty-vorpal'
import { Faq } from 'wechaty-qnamaker'
const vorpalConfig: WechatyVorpalConfig = {
contact : true,
mention : true,
room : true,
silent : true,
use: [
Faq(configCeibs),
],
}
const VorpalPlugin = WechatyVorpal(vorpalConfig)
const wechaty = new Wechaty()
wechaty.use(VorpalPlugin)
```
### 1 Configure QnAMaker
1. `endpointKey`: Endpoint Key for QnAMaker.ai
1. `knowledgeBaseId`: Knowledge Base ID for your knowledge base (KB)
1. `resourceName`: Resource Name for your Cognitive Service. (for example, `wechaty` is the Resource Name for `https://wechaty.azurewebsites.net`)
### 2 Language of Questions & Score of Answers
1. `config.language`: If set to a language ('chinese', 'english', etc), then the plugin will only reply message text in that specified language. (default: match all languages)
1. `config.scoreThreshold`: If the answer from QnAMaker.ai service has a score below the `scoreThreshold`, then that answer will not be used. (A perfect score is `100`)
### 3 Matchers & Skipper
1. `config.contact`: Whether to allow direct message to be sync with ticket reply. `false` to deny all, `true` for allow all; Supports contact id(`string`) and contact name(`RegExp`). You can also mix them in array.
1. `config.room`: The room id of your service WeChat room.
1. `config.mention`: Whether require the message mention the bot.
1. `config.skipMessage`: If set it to `string` or `RegExp`, then the message text that match the config will not be processed by the plugin. Array supported.
## ENVIRONMENT VARIABLES
The following two environment variables will be used if the required information is not provided by the config.
### 1 `WECHATY_PLUGIN_QNAMAKER_ENDPOINT_KEY`
`process.env.WECHATY_PLUGIN_QNAMAKER_ENDPOINT_KEY` will be used if the `config.endpointKey` is not provided.
### 2 `WECHATY_PLUGIN_QNAMAKER_KNOWLEDGE_BASE_ID`
`process.env.WECHATY_PLUGIN_QNAMAKER_KNOWLEDGE_BASE_ID` will be used if the `config.knowledgeBaseId` is not provided.
### 3 `WECHATY_PLUGIN_QNAMAKER_RESOURCE_NAME`
`process.env.WECHATY_PLUGIN_QNAMAKER_RESOURCE_NAME` will be used if the `config.resourceName` is not provided.
## EXAMPLE
Our Friday BOT are using `wechaty-qnamaker` to connect our WeChat conversations with QnAMaker.
1. Source Code - You can read the source code from: <https://github.com/wechaty/friday/blob/master/src/plugins/qnamaker.ts>
1. Spreadsheets - You can read our Question & Answer Pairs for Wechaty Community from: <https://docs.google.com/spreadsheets/d/14o7ytbZDRyX53nn8F4VohBgzAP6pMnCRPWFFzdcyzxc/edit>
> Note: our question & answer pairs sheet are open to edit. Please feel free to add question & answer pair if you believe it's necessary, and thank you for your contribution!
## RESOURCES
- [Quickstart: Test knowledge base with batch questions and expected answers](https://docs.microsoft.com/en-us/azure/cognitive-services/qnamaker/quickstarts/batch-testing)
## HISTORY
### master
### v0.6 (Aut 5, 2020)
1. Add Vorpal command: `faq`
1. Rename `minScore` to `scoreThreshold` in `configOptions`.
### v0.4 (July 23, 2020)
1. Rename from `wechaty-plugin-qnamaker` to `wechaty-qnamaker`
### v0.2 (June 29, 2020)
1. Init code base
1. QnAMaker integration
1. Add Language Detecter to limit the plugin to only answer a specific language(s)
## CONTRIBUTORS
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/0)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/1)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/2)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/3)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/4)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/5)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/6)
[](https://sourcerer.io/fame/huan/wechaty/wechaty-qnamaker/links/7)
## AUTHOR
[Huan LI](https://github.com/huan) ([李卓桓](http://linkedin.com/in/zixia)),
Microsoft Regional Director & AI MVP, \<zixia@zixia.net\>
[](https://stackexchange.com/users/265499)
## COPYRIGHT & LICENSE
- Code & Docs © 2020 Huan LI \<zixia@zixia.net\>
- Code released under the Apache-2.0 License
- Docs released under Creative Commons