From Peter's Blog: How it works
1. The search input is sent to wikipedia.org
2. The respective wikipedia page is sent to the Open Calais service to extract the terms
3. The extracted terms are sent to google, and get enriched by related terms using the google labs service "google suggest".
4. the terms are sent to the Amazon API and the relevant books from Amazon are displayed
A simple use case for this could be a blog widget that would semantically extracted information from posts, like for example on this post an Amazon widget would present books on the Semantic Web, Google and APIs with my affiliate ID embedded so if a reader wanted to purchase books related to the subject of the specific post they could.
1 comment :
I have been reading your blog for around 2 months, in a try to understand how this Semantic Web will work in practice but haven't managed to understand it yet. This post gives me a good opportunity to make another try though:
What's the difference of this search from the current ones that, let's say, Google does? Do we call it Semantic because we insert a word and the output varies from the wikipedia info, to related terms and suggested books from amazon? I mean, couldn't one build such a search engine even in Web 1.0?
Excuse me if I am missing something really obvious but I am some steps before hating this "web 3.0" thing, not because it's bad but because I don't understand it. And I really don't want to hate it, otherwise my Computer Science studies would find no field to be applied (yup, I want to become a web applications developer)... :)
Thanks for your time anyway!
Post a Comment