ApiAiRecognizer – 使用api.ai代替 Microsoft Bot 框架中的luis.ai
ApiAiRecognizer – 使用api.ai代替 Microsoft Bot 框架中的luis.ai
Dialogflow(API.ai) Recoginzer for Microsoft Bot Framework
Installation
npm install api-ai-recognizer
Usage
var builder = require('botbuilder'); var connector = new builder.ConsoleConnector().listen(); var bot = new builder.UniversalBot(connector); var apiairecognizer = require('api-ai-recognizer'); var recognizer = new apiairecognizer(
Using Entities
Entities can be found in the args.entities object and can be retrieved using code like below:
var city = builder.EntityRecognizer.findEntity(args.entities, 'city');
The schema for each entity retrieved with the about method is as below:
{ "entity" : "entity value", "type" : "entity name", "startIndex": "start index of entity", "endIndex" : "end index of entity", "score" : 1 }
Using Fulfillment
Dialogflow(api.ai) provides fulfillment which can be useful if you are using domains for your chat agent
Fulfillments are available as entitities with the entity name as fulfillment and can be used like below:
var fulfillment = builder.EntityRecognizer.findEntity(args.entities, 'fulfillment'); if (fulfillment) { var speech = fulfillment.entity; session.send(speech); }
Using Prompts
You can use prompts provided by Dialogflow(api.ai) for required entities very easily. Just check the actionIncomplete entity from the entities list and if its true use the fulfillment to send out the prompt.
var incomplete = builder.EntityRecognizer.findEntity(args.entities, 'actionIncomplete');if (incomplete){ var fulfillment = builder.EntityRecognizer.findEntity(args.entities, 'fulfillment'); if (fulfillment) { var speech = fulfillment.entity; session.send(speech); }}
Added functionaliy
onEnabled() function has been added.
Contributing
All feature requests, bug reprorts and pull requests are welcome!
License
MIT
版权声明:本文内容由网络用户投稿,版权归原作者所有,本站不拥有其著作权,亦不承担相应法律责任。如果您发现本站中有涉嫌抄袭或描述失实的内容,请联系我们jiasou666@gmail.com 处理,核实后本网站将在24小时内删除侵权内容。
发表评论
暂时没有评论,来抢沙发吧~