Semantic Web Explained |
Perhaps there is no better way of describing
the Semantic Web that to use the World Wide Web Consortium (W3C) explanations (http://www.w3.org/wiki/SemanticWebCerealBox)
Here's an explanation of the SemanticWeb by way of
analogy:
Let's say you're going to the grocery store. At the
grocery store, you get a box of cereal, right? So, you go to the self-checkout,
and shout to the computer, "I am buying a box of cereal!"
Of course, in this day and age, the computer doesn't
understand you. It just says, "Please slide the item across the
reader..."
So you find the bar code on the cereal. You slide it past
the laser reader. Suddenly- bingo. The computer knows what the item is, how
much it costs, how many you've bought so far, etc., etc., etc.,. Computers
don't yet know how to just "look" at the item, and know what it is.
So we make it easier for the computer: We tell the
computer in a language easy for it to understand. Every item in the store has
been given a number. That number has been correlated with other information in
a database. The number has been encoded into a bar code, because the laser can
read the bar code.
And the whole thing results in a computer that can reason
over a box of cereal that you're holding in your hand.
Resulting in faster and more accurate check-out.
Back to the Web
So, what's that have to do with the web?
The SemanticWeb is like bar codes for the web.
Say you visit your friend's web page. You can read all
this information on the web page, look at picture, etc., etc.,. But your
computer doesn't understand a thing about it.
If your friend wrote, "Hey friends! Call me up! My
number is 555-1212," your computer just sees it long stream of text. Sort
of like: If you write a letter to your friend about cats, your computer still
doesn't understand a thing about cats.
But now, let's put your friend's page on the Semantic
Web. Following the cereal analogy- let's make a little bar code tag, and
connect it to your friend's web page. Now, when you see that web page, you can
look for the attached bar code tag. In SemanticWeb terms, the bar code tag is
written in RDF. When your computer finds the RDF, it can read out all the
information.
Suddenly your computer knows your friend's name, what his
phone number is, who his friends are, etc., etc., etc.,. Maybe this all appears
in your address book. Or maybe you discover friends with similar interests.
The SemanticWeb will completely revolutionize the way
that we use computers.
At an ever faster pace, the semantic web is
beginning to influence how the internet interacts with us.
A tablet automatically tells you the weather at
your location. When you change location, guess what? The weather report is
automatically updated to the new location. You train arrival times are also
automatically updated. The internet ’knows’ what you are doing and can make
decisions for you (for example to look up the right information for your
current location).
In public relations it goes much further. It
is why the Lisbon Theory is so important. The Lisbon Theory provides a ‘bar
code tag’ that allows computers to understand what is being said and its
relative significance. The element of the ‘bar code’ are:
"From the values perspective (v) of an
entity (n) to what extent (e) is this object (o) significant (s)”.
If each of these elements are used to tag
semantic concepts using Latent Semantic Analysis (LSA), very significant
developments are available to the profession.
One of the joys of using LSA is that it
identifies highly linked concepts which, PR academic Bruno Amaral showed, have
close alignment to the values people hold. LSA is an approach to computerised
content analysis. In this it helps us manage relationships in an age of BigData.
With this information, the computer can get to
know what the content is about in public relations terms.
In addition, these technologies can learn from
human beings how they assess content from these five elements and learn to
emulate the human reaction to such content.
This means that, for example a human can teach
a computer to evaluate a corpus of media citations (press clips if you are still
of the old school). This will mean that
it comes much easier to follow those Tweets ’in bad taste’ as they happen or
identify good and bad content in blog posts and Facebook exchanges.
Before this book is too old to count, it will
also mean that a computer could respond online in near real time too!
At the time of writing, research in this area
is ongoing but gives us a view of what the potential of the semantic web brings
to public relations.
No comments:
Post a Comment