Waston NLC Hapi.JS starter kit

A simple Hapi.JS server that provides API endpoints to communicate with a [IBM Watson™ Natural Language Classifier service] (http://www.ibm.com/watson/developercloud/nl-classifier.html)

Getting started

You need a Bluemix account. If you don't have one, sign up. Experimental Watson Services are free to use.
Once you've created your Bluemix account and have logged in, create a new instance of the Natural Language Classifier to get the credentials for the service.
Clone or download the project, then install the dependencies by running the following node command npm install
While the dependencies are installing you'll need to get the credentials for your natural-language-classifier service.

  {
   "natural_language_classifier": [
      {
         "name": "natural-language-classifier-standard",
         "label": "natural_language_classifier",
         "plan": "standard",
         "credentials": {
            "url": "<url>",
            "username": "<username>",
            "password": "<password>"
         }
      }
   ]
}

Either add 3 new local env variables based on the props above, to your .bash_profile or enter them into your current working dir through terminal or command line. This gives us access to our new NLC service.

VCAP_NLC_URL=<url>
VCAP_NLC_USERNAME=<username>
VCAP_NLC_PASSWORD=<password>

Create and train the NLC service using, for example, the weather training data. Note the value of the Classifier ID in the response.
Start the server by running the following node command npm start
Point your browser to https:localhost:8000

API Endpoints

:id refers to the classifer id you created in step #7 above. You'll be able to easily obtain your classifer id using our list endpoint.

I've also named the endpoint and params to match how I use them on a daily basis.

List all classifiers

Returns a json object of all classifiers created within your service.

URL

/nlc/list
Method:

GET

Success Response:

Code: 200
Content:

  {
    "classifiers": [
      {
        "classifier_id": "10D41B-nlc-1",
        "url": "https://gateway.watsonplatform.net/natural-language-classifier/api/v1/classifiers/10D41B-nlc-1",
        "name": "My Classifier",
        "language": "en",
        "created": "2015-05-28T18:01:57.393Z"
      }
    ]
  }

Error Response:
- Returns an empty array if no classifiers are available.

Sample Call:

  $.ajax({
    url: "/nlc/list",
    dataType: "json",
    type : "GET",
    success : function(r) {
      console.log(r);
    }
  });

Ask a question

URL

/nlc/ask?:id&:text
Method:

GET
Query Params

Required:

id=[string] question=[string]

Success Response:

Code: 200
Content:

  {
    "classifier_id": "10D41B-nlc-1",
    "url": "<url>",
    "text": "How hot will it be today?",
    "top_class": "temperature",
    "classes": [
      {
        "class_name": "temperature",
        "confidence": 0.9998201258549781
      },
      {
        "class_name": "conditions",
        "confidence": 0.00017987414502176904
      }
    ]
  }

Error Response:
- Code: 404 NOT FOUND
  Content: { "description": "Classifier not found." }

Sample Call:

  $.ajax({
    url: "/nlc/ask?id=A3FCCBx16-nlc-412732?question=How hot will it be today?",
    dataType: "json",
    type : "GET",
    success : function(r) {
      console.log(r);
    }
  });

TODO

Finish documentation of api endpoints (Looking into Lout or Swagger)
Add 100% unit test coverage
Set up Travis CI

dfrsol / watson-nlc-hapi