Search Engine Comparisons - Pennsylvania State University

Download Report

Transcript Search Engine Comparisons - Pennsylvania State University

Search Engine
Comparisons
By: Thomie Ventura
Search Engines




Today, much, but not all, of the work we do
revolves around the web
Internet is accessible to almost anyone
Impact on businesses, schools, professionals,
home users
Web is changing every day, but everything is still
not ACCESSIBLE
FTP Servers



Only way of sharing files up to 1990
FTP Servers and FTP Clients
Down Side
Servers were mostly known through word of mouth
 Not everyone was setting up their servers

Grandfather, Grandmother, Mother

Archie ( Grandfather)


Veronica (Grandmother)


Used FTP file Servers
Used Gopher file Servers
World Wide Web Wanderer (Mother)
First Robot
 Caused Controversy


Are Robots a good or bad thing for the Internet?
“Web Search”

What exactly does it mean?
Involve tools ?
 Accessing proprietary databases such as
www.Factiva.com or www.dialog.com


We’ll focus on “web search” as an open web
source, and look at a searchers point of view
Difficulty Coping

Volume and Speed of the web and Search
Engines
Something new happens each day
 So many things to do, so little time to do it

Dynamic nature of web searching (indexing new
documents)
 Staying up-to-date with traditional tools( also undergo
changes)
 Other random issues that arise everyday

Will an “open web” search engine
always have my answers?

Questions that should arise about searching the
web
How long did it take to get it?
 What is the database or search engine?
 What kinds of questions will it help me answer?


Open web will not always give me the answer

What can it be used for?
Quality of Information


Anyone can become a publisher
Evaluating content is crucial
Reputation
 Background
 Qualifications
 Where did it come from?
 What its purpose?
 Relevant to my topic?

Limitations of General Web Search
Tools

Spiders don’t crawl in real-time


Recency
Linked or Submitted Sites

If a website contains 1000 pages, does not mean
Search Engines make all of them accessible
Invisible or Hidden Web resources

Examples:
Interacting resources, return “custom” sites
 Registration


Why is it hidden?
Created on the fly
 Spiders don’t fill in registration forms
 “No-Robot” Tag

Hidden is not always bad


Research and Effort
Without proper tools, we can make large
databases even larger
Google
 Altavista
 Excite


Distributing Information Properly
Specialized Focused and Site Specific
Search Tools

Necessary and Important



Hidden Web is out of reach of general purpose
Search Engines
More Precision than Recall
Examples:

www.Psychcrawler.com www.Inomics.com
[http://newssearch.bbc.co.uk/
ksenglish/query.htm],
Identifying and Collecting
Specialized Engines

Profusion


[http://www.profusion.com]
Librarians Index
Covers large amount of specialized and invisible web
databases
 [http://www.lii.org]

Meta – Search Engines

Major Disadvantages
You get it all!! High Recall Low Precision
 Basics of Search Engines used
 Send queries to “pay for placement” engines


A good metasearch Engine

www.vivisimo.com
Old Pages, GONE!

Trying to find old pages?


Contact webmaster
Fortunately
Archiving Old Material
 Example:

[http://www.clinton.nara.gov/index.html]
 ALexa Research



[http://archive.alexa.com/]
carries over 18 terabytes of data covering some 5 million Web
sites and some 1.9 billion pages
Search Engine Sizes


This is a search engine
size analysis as of
December 11, 2001
Google Dominates
Sizes Over Time
Closer Look
Dealing with Coping

Use the Search Engine

Conduct research on a topic
This will get you familiar with search engine
 You can see how results are displayed



Relevancy of returned documents
Let you gather your own bookmarks
Understanding limitations

What to do with these limitations?
Know limitations
 Use more than one search engine
 Use “specialized” search engines that go deeper into
a site to collect more information
 Use “invisible web” resources
 Use web directories, and bookmark important sites

Ability to Search Multimedia

Now Available, but still expanding
Wait weeks now becomes instant
 search tools that provide access to video and audio
material using a non-text mechanism to access the
material ex: searching a specific background or type color
 Still image tools


Google, Altavista, and Fast, use text surrounding image
Become Aware of Multimedia Search

Video Searches






Virage www.virage.com
TVeyes www.tveyes.com
ShadowTv www.shadowtv.com
Wordwave www.wordwave.com
SpeechBot (keyword search engine demo by Compaq, uses speech
technology to create real-time transcripts) www.speechbot.com
Image Searches


Webseek (search or browse criteria in image)
www.ctr.columbia.edu/webseek/
Visoo( uses software that looks for words embedded in image
www.visoo.com
Making Old Pages Stay

Long Term?


Offer comments ( suggest how material can be more
accessible and searcheable, a great archive of content without
the correct means of accessing it will be a hassle and is not
great)
Short Term?


Take advanatage of Googles cache feature ( google crawls a
site and makes a copy unless unauthorized, and puts it on
server, if site is gone, the copy is in googles server, you must
go to search results and next to URL go to “cached”, will not
always be there, next time spider crawls site and it is missing
it will not save onto server
www.savethis.com (lets you save web pages, and access them)