{"id":3272,"date":"2015-01-25T20:18:54","date_gmt":"2015-01-25T20:18:54","guid":{"rendered":"http:\/\/dalelane.co.uk\/blog\/?p=3272"},"modified":"2015-01-26T18:39:58","modified_gmt":"2015-01-26T18:39:58","slug":"how-to-use-the-ibm-watson-relationship-extraction-service-on-bluemix","status":"publish","type":"post","link":"https:\/\/dalelane.co.uk\/blog\/?p=3272","title":{"rendered":"How to use the IBM Watson Relationship Extraction service on Bluemix"},"content":{"rendered":"

Before Christmas, I wrote about how I used the Watson Relationship Extraction service on Bluemix to pick out the things mentioned in news stories<\/a>, as part of a mobile app we built on a hackday<\/a>. I’d still like to do something more with that app, but in the meantime I should at least share how I did the Relationship Extraction bit. <\/p>\n

From the official doc for the service<\/a>:<\/p>\n

From unstructured text, Relationship Extraction can extract entities (such as people, locations, organizations, events), and the relationships between these entities (such as person employed-by organization, person resides-in location).<\/p><\/blockquote>\n

This is provided as a hosted service on IBM Bluemix<\/a> where any developer can sign up and give it a try. <\/p>\n

It’s available as a documented REST API<\/a>, but as part of using it in the hackday, I needed to write a bit of code around that, just to prepare the request and parse the response. I think it’ll save me time to reuse this the next time I want to build something with the API, so I’m sharing it as a standalone package<\/a>. <\/p>\n

In this post, I’ll walk though how you can use it, with a small app that grabs the contents of a BBC News story and picks out the names of people mentioned in the story. <\/p>\n

First, a simpler example. Consider this exciting text:<\/p>\n

Dale Lane works as a developer for IBM. He started in 2003. Dale lives in the UK, in a town called Eastleigh. Before that, he was a student at the University of Bath.<\/p><\/blockquote>\n

A few lines of Javascript are enough to run that through the service. <\/p>\n

var text = 'Dale Lane works as a developer for IBM ... ';\r\nvar watson = require('extract-relationships');\r\nwatson.extract(text, function(err, response) {\r\n    \/\/ response has got all the info  \r\n});<\/pre>\n
The full contents of response<\/code><\/a> is in a gist if you want to see it, but I’ll show just a few examples here to give you the idea. <\/p>\n
screenshot<\/p>\n
It has picked out all of the references to me, recognising that they are all describing a person, and that ‘developer’ is my occupation. <\/p>\n
The ‘begin’ and ‘end’ numbers tell you where in the text each bit was found. <\/p>\n
screenshot<\/p>\n
It’s picked out the reference to IBM, and recognised that this is a name of a commercial organisation. <\/p>\n
screenshot<\/p>\n
It’s recognised that ‘2003’ is a reference to a date. <\/p>\n
As well as identifying those entities and many others, it’s also picked out the relationships between them. <\/p>\n
screenshot<\/p>\n
For example, it’s identified the relationship between me and IBM. <\/p>\n
screenshot<\/p>\n
And the relationship between me and my old University. <\/p>\n
I’ve written a more detailed breakdown of what is contained in the response<\/a> including how to find out what each of the fields mean, and what the different possible values for each one are. <\/p>\n
That’s the basics with a few input sentences. Next, we start throwing a lot of text at it. <\/p>\n
In about thirty lines of Javascript<\/a>, you can download the text from a news story on the BBC News website, and pick out the names of all of the people mentioned in the story. <\/p>\n