Davey Winder
Friday, 04 July 2008 04:33
Your IT -
Home IT
Page 2 of 3
How does Microsoft, and its acquisition of Powerset, fit
into the semantic web road map? Well, it will argue that today roughly
a third of searches are not answered effectively on the first search or
the first click. Multiple searches and multiple search results work for
most people most of the time, but they are far from being perfect.
In searching for perfection, Microsoft wants to
adopt a semantic web approach to the problem of differences in phrasing
or context between a search terms and web data. To get around the huge
chasm between human understanding of linked concepts and the inability
of the web to do the same.
"These problems exist because search engines today primarily match
words in a search to words on a webpage" Nadella insists "We can solve
these problems by working to understand the intent behind each search
and the concepts and meaning embedded in a webpage."
Which is where Powerset comes in with its natural language processing
expertise. It already has the technology to parse natural language
questions and return topic clusters with a contextual base.
Although Microsoft nor Powerset are saying how much money was on the
table, in the last few months online rumour has abounded which seemed
to focus on the figure of USD $100 million (AUD $104 million). Some
are
now reporting
this figure as fact. If this turns out to be anywhere near the right
ballpark, then it seems to me that Microsoft has got itself a bargain.
Let's face it, anything that can help Live Search to both parse better
at the front end and cluster results better under the hood has got to
be money well spent.
I think
John Blossom has hit
the nail on the head though, when he asks why if Microsoft is so
interested in natural language search it didn't buy Answers.com
instead? "With millions of pre-formed questions already in its
WikiAnswers database many natural language questions map very neatly to
its answer sets" Blossom argues "In other words, sometimes the best
answer to a full-sentence is a person who understood the question in
all of its semantic details and has already provided the answer."
Does this mean that Microsoft will bounce back to bite Google in the search market? Read on to find out...
CONTINUED