Transcript Semantic Web and the Perfect Search

Semantic Web and the
Perfect Search
By: Nolan Smith
Semantic Web
• Tim Berners-Lee started semantic web
around 1998 or so, and quickly gained the
attention of an increasing circle of people.
• The SemWeb was originally supposed to
give the web the “smarts” it lacked and
much of the early work on it dealt with
calendaring, scheduling, and expressing
relationships between people.
Semantic Web (cont’d)
• Purpose: provide an easier way to find
things on the web.
• Executes complex tasks built upon finding
things such as
– scheduling a meeting
– planning a trip
– organizing a wedding
– Etc.
Semantic Web (cont’d)
• Example from reading:
– If A is a friend of B, then B is a friend of A.
• Prof Babu has a friend named Nolan
• Therefore, Nolan has a friend named Prof. Babu
• So if I find a statement on Prof. Babu’s
web site that says, “Prof. Babu is a friend
of Nolan” and someone does a search for
Nolan’s friends, even if Nolan’s web site
doesn’t mention Prof. Babu, we know Prof.
Babu considers himself a friend of Nolan.
Search for Perfection
• Rise of Semantic Web: the tagging of
information so as to make it more easily found.
• Other Improvements:
– Ubiquity: the integration of more and more information
into the web indexes.
– Personalized search: the application of your personal
Web toward a more perfect answer.
– Domain specific search.
– Web time axis.
Ubiquity
• Critical to perfect search
– Means nothing if the search engine does not
understand your likes and dislikes.
• Gives every piece of information, such as music
or books, and index which makes it easier to find
• Example
– Napster: millions of people ripped copies of their
favorite music to the Web.
– It became much easier to find any song on the Web
Personalized Search
• Local search = personalized search
• Google’s version of local has two inputs:
– The search term itself
– A bit of local information (i.e. zip code or town
name)
• Example
– Searching for basketball tickets
• Search will return tickets in Durham not Russia
Domain specific search
• Battelle states “Domain specific search solutions
focus on one area of knowledge, creating
customized search experiences that, because of
the domain’s limited corpus and clear
relationships between concepts, provide
extremely relevant results for searchers” (274).
• Domain specific search only searches for things
within a specific area, such as cars or computers
• Example based on what we’ve learned about
Ubiquity and Domain Specific Search
– When I search for “Jaguar”, the search engine knows
to return the car and not the animal
Web Time Axis
• Allows for a search constrained by date
– Specific time period instead of searching for
specific dates.
• You could ask the search engine
– “Show me all results for my query from a
specific time period”
– “Tell me what were the most popular results
for ‘George W. Bush’ on May 3, 2004”
The End!!