Finally it’s time for an update of our text analysis demo site, not a day to soon! Our old demo site was made in a rush, I think it was right before a conference of some kind. Back from the conference we figured that since the demo seemed to be working well enough, we could leave it for a couple of days before starting to improve it… That was over a year ago.
The goal with our new text analysis demo is more then anything to make it easier to try out. That’s why the frontpage at demo.saplo.com now has a predefined text in the textarea. That way you as a visitor can just click submit and get results for the text immediately.
You can also toggle between English and Swedish texts or you can switch between news articles and tweets. If that doesn’t do it for you, there’s also the option of writing or pasting your own text into the textarea.
The demo provides you with three different kinds of results:
- Entity Tagging
- Similar Articles
- Similar Tweets
Give the demo a try at demo.saplo.com!
How is the demo built?
First of two different text collections is created through the Saplo Text Analysis API using the collection.create method. The first collections store articles and the second tweets.
We crawl news articles from a couple of news sites and we collect tweets from specific Twitter streams. Tweets and articles are added to the two different text collections using the API method text.create.
Next we create a third text collection, in this collection articles and tweets written by visitors will be stored.
That’s it, nothing more fancy about it.