IBM Watson - Natural Language Understanding

• Natural Language Understanding uses natural language processing to analyze semantic features of any text. Provide plain text, HTML, or a public URL, and Natural Language Understanding returns results for the features you specify. The service cleans HTML before analysis by default, which removes most advertisements and other unwanted content.

Pre-requisites

• We will use Node.js along with the express framework to build our server. Ensure that you have Node installed on your computer, do this by running node-v in the terminal. If you do not have Node installed you can get it from https://nodejs.org

• Before you begin, ensure that you have the latest version of the hasura cli installed. You can find instructions to download the hasura cli from here

Doing the basic setup in IBM Watson

• Navigate to https://console.bluemix.net/registration/ to register on IBM.

• Log in to IBM Cloud and go to the Natural Language Understanding service page.

• Click the "Create" button.

• Now head over to dashboard after you have created the service.

• Click on the service name in the dashboard which will take you to the NLU Service page.

• Click the "Service Credentials" tab on the Natural Language Understanding page in your IBM Cloud dashboard to view your credentials.

• click on View Credentials to see your credentials to be used further.

{
  "url": "https://gateway.watsonplatform.net/natural-language-understanding/api",
  "username": "dh4il910-241a-49a8-lk90-bd7b83a185f1",
  "password": "avb5hdbswk"
}

Getting the Hasura project

$ # 1) Run the quickstart command
$ hasura quickstart hasura/hello-nodejs-express

$ # 2) Git add, commit & push to deploy to your cluster
$ cd hello-nodejs-express
$ git add . && git commit -m "First commit"
$ git push hasura master

After the git push completes:

$ hasura microservice list

You will get an output like so:

INFO Getting microservices...                     
INFO Custom microservices:                        
NAME   STATUS    INTERNAL-URL(tcp,http)   EXTERNAL-URL
bot    Running   bot.default              https://api.apology69.hasura-app.io

INFO Hasura microservices:                        
NAME            STATUS    INTERNAL-URL(tcp,http)   EXTERNAL-URL
auth            Running   auth.hasura              https://auth.apology69.hasura-app.io
data            Running   data.hasura              https://data.apology69.hasura-app.io
filestore       Running   filestore.hasura         https://filestore.apology69.hasura-app.io
gateway         Running   gateway.hasura           
le-agent        Running   le-agent.hasura          
notify          Running   notify.hasura            https://notify.apology69.hasura-app.io
platform-sync   Running   platform-sync.hasura     
postgres        Running   postgres.hasura          
session-redis   Running   session-redis.hasura     
sshd            Running   sshd.hasura              
vahana          Running   vahana.hasura

Find the EXTERNAL-URL for the service named api(in this case -> https://api.apology69.hasura-app.io).

Tutorial

Follow along for a step by step guide on developing and deploying this facebook messenger bot on Hasura.

Sections

• Architecture
• Getting started

Architecture

Getting started

Adding additional node dependencies

Navigate to microservices/bot/app/src (the directory with the package.json file)

$ npm install request body-parser --save

• The above command installs two modules(libraries) "request" and "body-parser" into our app.
• request is for sending out messages and body-parser is to process messages.
• Once again, open up the server.js and add the following at the top of your file:

    var bodyParser = require('body-parser');
    var request = require('request');

and the following after var app = express();

    // Process application/x-www-form-urlencoded
    app.use(bodyParser.urlencoded({extended: false}));

    // Process application/json
    app.use(bodyParser.json());

• Your server.js file should now look like so:

    var bodyParser = require('body-parser');
    var request = require('request');
    var express = require('express');
    var app = express();

    // Process application/x-www-form-urlencoded
    app.use(bodyParser.urlencoded({extended: false}));

    // Process application/json
    app.use(bodyParser.json());

    //your routes here
    app.get('/', function (req, res) {
      res.send("Hello World!");
    });

    app.listen(8080, function () {
      console.log('Example app listening on port 8080!');
    });

Getting started with IBM Watson - NLU service

Step 1 - Authentication

Replace {username} and {password} with your credentials which you generated in Service Credentials.

var NaturalLanguageUnderstandingV1 = require('watson-developer-cloud/natural-language-understanding/v1.js');
var natural_language_understanding = new NaturalLanguageUnderstandingV1({
  'username': '{username}',
  'password': '{password}',
  'version_date': '2017-02-27'
});

Step 2 - Making a POST call to the service to categorise the text.

app.post('/', function (req, res) {
    var inputString = req.body.input;
    compute(inputString);

    function compute(input) {
        var parameters = {
            'text': input,
            'features': {
                'categories': {}
            }
        };

        natural_language_understanding.analyze(parameters, function (err, response) {
            if (err)
                console.log('error:', err);
            else
                resp = JSON.stringify(response, null, 2);
            console.log(resp);
            sendRes(resp);
        });
    }

    function sendRes(output){
        res.setHeader('Content-Type', 'application/json');
        res.send(output);
    }
});

The above snippet does the following:

1. Authentication with IBM WATSON - NLU Service.
2. We a post route setup to which the front-end makes a call.
3. The function compute(input) is executed by the post request.
4. And returns a JSON in the response.

Example -

• The example response printed in the logs will look like so

    {
      "usage": {
        "text_units": 1,
        "text_characters": 1536,
        "features": 1
      },
      "retrieved_url": "https://www.ibm.com/us-en/",
      "categories": [
        {
          "score": 0.594296,
          "label": "/technology and computing/software"
        },
        {
          "score": 0.448495,
          "label": "/science/mathematics/statistics"
        },
        {
          "score": 0.426429,
          "label": "/business and industrial"
        }
      ],
      "language": "en"
    }