WEB Searching

1
WEB Searching

• Pritam Dhavale
• Subodh Kerkar
• Satish Bhat

2
World Wide Web

• Internet is an inter-connected network of
  networks

3
Who owns the Internet?

• No one owns the internet
• So users must carefully evaluate what they find
• Who is publishing the site and why?
• When was the site last updated?
• How relevant is it to our need?

4
Most common Searching Errors

• Incorrect splling
• Poorly described Query
• Query is too general
• Query is too specific
• Search syntax is wrong

5
Web Directory

• Is a subject tree or subject directory
• Its a listing of subjects organized into
  categories or hierarchy
• It can be organized in a variety of ways such as
  topically or alphabetically

6
Difference between a Directory and a Search Engine

• Directory
• - are compiled by people
• - rely on the judgment and
  expertise of
• the people compiling then
• Search Engine
• - are automated
• - can also be called as intelligent
  agents

7
Usage

• Directories
• - use it when you want to see what
  is
• available on a particular topic
• Search Engines
• - use it when time is not a
  crucial factor
• and when you would like to see
  more
• new information on a particular
  topic

8
Examples

• Directories
• - Argus Clearinghouse
• (www.clearinghouse.net)
• - Encyclopedia Britannicas
  Internet
• Guide (www.britannica.com)
• Search engines
• - Google (www.google.com)
• - Alta Vista
  (www.alta-vista.com)

9
Working of a search engine

• 3 Mechanisms
• Identification of the web pages
• Indexing of the websites and
• Searching mechanism with an interface
• User is exposed to the 3rd mechanism

10
Query strings

• Natural language querying or querying by asking a
  question
• Ex. www.askjeeves.com
• Loose words

11
Heard about them ?

• Intelligent agent,
• Spider,
• Crawler,
• Robot or
• Worm..

All are the same
12
Working of a robot

• They search terms matching a certain criteria
• Simultaneous work takes place
• Huge index is created

What is an index ???
13
Index

• It is used by the search engine to locate the URL
• A computerized book index example

14
Example
15
Contd

• A search engine uses its index to retrieve web
  documents in which the search term occurs
• A string-URL pair
• Each search engine has its OWN index

16
Relevance

• What is it ?
• Probability that the match is on-target
• Its calculations is confidential
• Weight of words in key positions

17
The downside

• Spamming of a search engine
• Invisible text
• Repeated use of words in meta-tags
• Artificial ratings
• Selling of higher ratings

18
When to use which engine
19
Power Search Techniques

• Why use power search?
• Time-consuming
• Not very precise
• What are the power search techniques?
• Boolean searching
• Field Searching

20
Boolean Searching

• Implementation of Boolean Logic and set theory
• AND, OR , NOT, ADJACENT, ALL, ANY and NEAR
  options
• , - options

21
Example The NOT query

• Information about cats but nothing about dogs

22
Field Searching

• Specifies where to look in the web document
• For example
• in the title only, or the URL fields
• Powerful tool as it gives more control to the
  user

23
Image Searching

• Images are searched by
• filename extension
• like .jpg or .gif
• Html attributes
• like IMGSRC, HREF

24
Meta-search Engines

• Queries various search engines in parallel with
  the keyword and displays the results
• Example Metacrawler, Dogpile
• Major Differences amongst meta-search engines are
  with respect to
• databases queried
• response time

25
Conclusion

• Google has started a new service that lets Web
  developers perform automated searches of its vast
  Internet database and publish the results on
  their own sites
• Voice recognition