GO HOME
Make Start Page
Tell A Friend
Tips For Kids
↓NAVIGATION BAR↓
GENERAL
Directory Search
General Search
Meta Search
News Groups
BUSINESS
Administration
Legal
Marketing
Money & Stocks
Shipping
COMPUTERS
AntiVirus
Downloads
Hardware
Internet
Software
Support
GEOGRAPHIC
Maps
Travel
Zip Codes
HOME & HEALTH
Art & Photos
Books
Children
Cooking
Health
Magazines
Music
Shopping
Video
REFERENCE
Acronyms
Dictionary
Encyclopedia
Translations
TELEPHONE
Area Codes
People Finder
Reverse Directory
White Pages
Yellow Pages
MEDIA
News
Sports
TV & Radio
Weather
| Advanced Internet Search Techniques
|
|
| Search engines are software tools designed to help you find information on the Internet. No individual search engine has yet examined or indexed the entire World Wide Web. This is not so much because it is impossible to do, but because it's impractical, especially considering performance, and the unmanageable number of results that would be returned from any query. Consequently, each site contains only a partial amount of the total information available, which is then organized, dependent on the kinds of information the engine gathers and classifies. Generally, search engines fall into one, or more, of the following categories:
| | | Many of the most popular search engines on the Web are known as Indexing Engines, sometimes called “Crawlers” or “Spiders”. These names are derived by extending the “spider web” metaphor to the actual way Indexing Engines find information on the Internet. Using software that scans (“crawls”) through Web pages, these engines gather, classify and index pieces of information based on a pre-determined set of criteria. (primarily keywords, their location on a page, combined with meta-tags, web links, etc.) The indexed information is then automatically placed into the engine's database, which is searched each time you query the site.
Since Indexing Search Engines have such broad and diverse databases of web links, you should expect many results, even if the information you want is extremely hard to find. Be careful though, if your query is too simple you are likely to get hundreds of thousands of results, with no guarantee your best information is anywhere near the top of the list.
Here are some popular examples of Indexing Search Engines: | | | Directories are categorized groupings of sites, most often compiled and organized by real human beings. They are organized into a series of categories and sub-categories, moving from the general to the specific. Each sub-category yields a list of additional sub-sub-categories etc., etc., etc., until finally a list of web sites results. While the quantity of results are usually much fewer, and the search process is a little longer than an Indexing Engine, the relevancy of results to your topic is usually higher.
For ease of use, most Directories also have a search feature, which can actually be a little misleading. This “search” function enables you to search their listings. Remember, you will not be searching the Internet. Instead, you will only be searching that Directory's categories and listings (i.e. titles, descriptions and URLs (web site addresses) as they appear in their database, not the contents of the sites themselves. NOTE: Sophisticated keyword search strategies are of limited use here.
You can use Directories to good advantage when narrowing down a broad subject, or when you need general information on a range of topics.Top Of Page Some examples of Search Directories are: | | Natural Language Search Engines aren't really of much use to advanced users. Mention is made of them here solely for the purpose of providing complete coverage on the subject of search engines. For beginners, or for people who prefer to "ask" simple questions rather than using keywords and logical operators, they are a comfortable alternative.
On a Natural Language Search Engine, the user submits a query in the form of a simple question, which usually returns results in the form of other questions which help narrow and clarify the search. While this is not an efficient way to find high quality information, it is a much better strategy than submitting a single keyword to an Index Search Engine, which happens all too frequently with inexperienced users.
Top Of Page An example of a Natural Language Search Engine is: | | | Search engines are the master keys that make possible the broad-based, egalitarian Internet we presently enjoy. Without them, access to online information would be available only to a privileged few, who even then could only retrieve documents whose filenames and storage locations were known to them. As important as search engines are, the very fact they exist flies in the face of conventional business wisdom. Consider their typical business model; a non-governmental entity, that costs huge amounts of money to establish and maintain, whose product is provided free to users and providers alike. In other words, not highly profitable. Top Of Page So, in an effort to improve profitability, the Pay Search Engine was born. When users first encounter the term, “Pay Search Engine”, many jump to the conclusion that they will have to pay something to make use of them. Of course, nothing could be further from the truth. The word “pay” here applies only to web site owners who want their sites listed on the search engine's database. The names of the types of pay engines corresponds to the way in which web site owners are charged for the service.
The two principle types of these engines/programs are Pay-For-Inclusion and Pay-For-Placement (also called Pay-Per-Click). Pay engines are quite attractive to owners of income producing sites for several reasons;- the application-for-listing process is faster and less complicated than traditional methods
- the successful outcome of that process is more certain
- the cost of the listing can be offset through incremental revenue that is roughly proportional to site performance
It's easy to see why these engines/programs have been steadily gaining in popularity among owners of commercial web sites.Top Of Page Pay-For-Inclusion engines typically require an up front, fixed fee to be considered for inclusion in the search engine's database. Placement within any generated query results is not guaranteed, nor is the overall ranking within those results. Pay-For-Inclusion sites usually also offer an alternative, free submission program, but the process takes more time.
Some examples of search engines with Pay-For-Inclusion programs are:Pay-For-Placement (or Pay-Per-Click) engines usually do not have fixed rates. Web site owners pay on a CPC (Cost-Per-Click) basis, where they pay a certain amount every time someone clicks on a link to their site. The CPC is a negotiated price, and is established in an almost auction-like environment where owners try to outbid each other to get their sites listed higher in the query results.
Some examples of Pay-For-Placement Search Engines are:One final note on Pay-For-Inclusion and Pay-For-Placement engines; since web site owners are paying for the listing, their sites are not likely to be great sources for free information. The vast majority of the listings in this category are commercial sites. i.e. retailers, online stores, etc. These web sites are most useful when you are looking to buy something. | | | For the most part, meta-search engines are sampling engines. Rather than search a database of their own, they query a series of other search sites simultaneously. After compiling the results (maybe removing duplicates), they display them in a way that varies from engine to engine. Some may combine all results in a ranked list, while others separate them by source.
Meta-search engines are generally more tolerant of imprecise search terms and operators than other types of engines. They gather only the top few results from many search sites, so they tend to return fewer results with a greater degree of relevance. This kind of approach allows the user to get a kind of snapshot of the top results from many different search engine kinds and types (indexes, directories, pay-for-placement, etc). It's a great way to get a quick overview of the kinds of information that are available on your subject of interest. Also, if you are in the habit of using a “favorite” search engine, you can use a meta site to make sure you haven't missed any great resources available elsewhere.
Some examples of Meta-Search Engines are: | | | The World Wide Web is growing at a faster and faster rate every day. Owners of search engines realize they cannot index or categorize all of it, so in order to maintain the usefulness and profitability of search engines, a trend of cooperation and partnership between search engine types is on the increase. Consequently, the boundaries between types of search engines is getting less defined, and this trend seems unlikely to change.
Sometimes, when a search engine's database is limited on a subject, they get another search engine to serve results when they've run out or don't have any of their own to display. This is often the case with Directories, since their overall coverage is more limited than Indexed engines.
For example: When searching Yahoo (a Directory), and their resources are exhausted, you will automatically be provided with Google results (an Indexing Search Engine). Much of the results from MSN consist of the LookSmart Directory, and even LookSmart employs Inktomi to serve backup results for them, and on and on.
Some Indexing Engines, such as Lycos and AltaVista, serve Overture results (Pay-For-Placement) above their own. While all this may sound a little complicated and confusing, serving backup results actually makes good sense, because it provides a more positive user experience, since they will seldom encounter search returns with very few, or no results.
So, the results you get from your query can be coming from sources other than the ones you think you're searching. Where they come from may, or may not, be differentiated on the results page. Since you ARE getting information this may seem an insignificant point, but it is important to pay attention to the details, and to know where your results originate.
(see “Evaluating Your Results”)
| |
| | | | There is an enormous amount of information stored on the World Wide Web. This is undeniable. Just about any topic you can imagine is covered in detail, and from every point of view. That being said, how do you get the information in which you are intersted, from the perpective you want?
The best way to begin any project is with a plan, and in this case the plan will take the form of a search strategy. In order to develop this strategy, a few questions must first be answered.- What kind of information is required? i.e. articles, pictures, software, etc.
- Will general information be adequate at this time, or must it be more focused?
- How current must the information be?
- Is something very specific required? i.e. printer driver, current news photo, magazine article, etc.
- Are there any groups, companies or organizations who are known to possess the required information?
- What type of search engine will likely yield the best results for this kind of information?
- What words are likely to be associated with this subject?
- What other questions are appropriate to this search?
The better you can define the information you are looking for, the more probable it is you will find authoritative sources containing it.
Top Of Page Advanced techniques are great, but always remember the fundamentals:
- Spelling: One of the most common mistakes made when searching for information is misspelling. Many times this can be the result of typographical errors caused by entering keywords too quickly from the keyboard, or just not being sure of the correct spelling. Keep in mind that the search engine will match your results EXACTLY to your query. You may encounter some interesting results with a word spelled wrong, but most often nothing is returned. Sometimes, it is worth considering alternative spellings, especially when correct alternative spellings exist, as with American and British English. (e.g. honor-honour, color-colour, program-programme, etc.)
- Word Stems & Endings: Consider ALL spellings and forms of words, including plurals, verb tenses and gerunds. Different results will be returned for cat and cats, or run and running, or for parents and parenting. Do not assume you've performed a thorough search until you've explored these kinds of possibilities.
- Synonyms: Using different words that mean the same thing will also change your results. For instance, consider searching for both “heart attack” and “cardiac arrest”, or TV and television, or “cancer specialists” and oncologists.
Top Of Page - Be Specific: Using precise terms will always yield the most relevant, if fewer results. For instance, if you live in Chicago, and you're looking to buy a Subaru Outback, you wouldn't want to search for “Subaru Outback” or “station wagon”. Instead, a query for Subaru dealership in Chicago would yield better results.
- Make The Search More General: If being specific is returning too few, or no results, it is probably necessary to broaden your search. This is especially true when your subject of interest is already very specialized in nature. Searching for general categories related to your topic will return more results with generally less relevance, but at least you will have some information. Once you've found out what kind of information is available, you can refocus your search.
The type of information you're trying to find will affect how you go about looking for it. Read through the descriptions of the various search engine types (see Guide To Internet Search Engines) to find out which one best suits your needs. Once you've determined that, in most cases the next step will be to perform a “Keyword” search.
| | | Basic Keyword Searching
Unless you are using a Directory (such as Yahoo or LookSmart), or a Natural Language Search Engine (like Ask Jeeves), you will be using keyword terms with logical operators to search the Internet. Here are a couple basic tips to help you get better results when using keywords to search the Internet with the “Little Engine”. While nothing works exactly the same on all sites, these techniques work well on most.- A really easy way to narrow your search is to enter several keywords, separated by spaces, to indicate you would prefer documents containing all of the words. Click the example below.
Example: chicago restaurants italian
Top Of Page - You can also use quotation marks or a dash without spaces to indicate you are looking for the occurrence of an exact phrase.
Just remember to be specific. Click on the following two examples to see the difference in results.
Example: "Ulysses S Grant" or Ulysses-S-Grant
Example: "US Grant" or US-Grant
| Problems with Keyword Searching
The main problem with searching using keyword terms is the difficulty search engines encounter when trying to determine relevance. While the words you use in your query have meaning to you, search engines see them only as a string of characters.
Many strategies are employed by the various search engines to address this issue, all with mixed success. Some engines determine relevancy according to how frequently the keyword occurs in the document, or where it appears on a page. Neither of these ideas translate to good, relevant information consistently. After all, you're looking for articles written ABOUT your keywords, not ones that repeat them frequently.
Top Of Page Since the search engine sees your keywords only as ASCII strings of characters that it must match, they have no way of knowing that when you are looking for "TV listings", you might also be interested in "television listings". To make up for these shortfalls, users must think through the search process a little better, and modify their queries accordingly, to get the best results.
Most search engines handle your query by first retrieving results containing all of your search terms. When it runs out of documents that meet that criteria they gather those that contain some of your terms, and so on, until no documents match. Then the results are sorted by whatever logical scheme the engine uses to determine relevancy, and displays them in a list. Results displayed toward the end of that list are probably very irrelevant, since only one of your terms found a match.
To avoid the overhead of building extremely large lists, and to present the user with more useful information, most search engines support some type of advanced search features. That way you can be more specific in your search, telling the search engine EXACTLY the kind of results you desire. The following types represent the most broadly supported of these operators and techniques.
| | | Search Math
Search math, also known as Advanced Search Syntax, is now available on all the most popular search engines. The symbols used correspond to the four (4) basic types of query syntax:- Required and Prohibited Search Words
- Document Section Restrictions
- Phrase Matching
- Wildcard Matching
Required and Prohibited Search Words(+ -)
Attaching one of the following operators will either require or prohibit words from appearing in the search results.
+ requires that all search results contain that word.A word of caution: using the + sign does not mean that your required word will be present in the same sequence as they appear in your query. It only guarantees that the word will be on the page somewhere.
- requires that NO search results contain that word.Also, take note of how little attention is paid to the terms of your query by the "Sponsor" Matches. It's as if you gave the opposite instructions.
Top Of Page A note on Stop Words:
Many search engines have a predetermined list of words they ignore, regardless of your search operators. These are called Stop Words. Examples include or, of, a, an, in, on, to, the, what, etc. A few search engines are more friendly about this, so if you think your query requires the inclusion of stop words, go ahead and use the Required and Prohibited operators. It won't hurt anything to try.
Document Section Restrictions(t: u:)
Attaching one the following operators to the front of a search word will restrict the search to a certain document sections.
t: will restrict searches to document titles only- Example: joe boxer versus t:“joe boxer”
u: will restrict searches to document URLs only- Example: intel versus u:intel
Top Of Page Phrase Matching (“ ”)
“” will match all the words in the exact sequence.- Example: great barrier reef versus “great barrier reef”
Wildcard Matching (*)
* will return complete word matches for the word fragment and unknown letters on the side of the * symbol.Top Of Page Combining Syntax (+ - t: u: “” * )
You may combine any of the query syntax symbols as long as the syntax is combined in the proper order. The proper order for using the syntax is the same order that the operators are presented above. That is, +, -, t:, u:, “” and lastly *. Also, remember to leave a space between individual operator/keyword groups.- Example:
(correct) +t:football -American
versus
(incorrect) +t:football-American
or
(incorrect) t:+football -American
| | | Boolean Operators:
Performing much the same functions as search math, Boolean logical operators also allow you to specify what you want included in your results. Here are the ones most used:
AND
The Boolean AND is the word equivalent the + sign. It specifies that all terms submitted MUST be present on the returned pages.
OR
Search math doesn't have an OR operator. It specifies that at least one of the terms submitted must occur on the pages returned. Boolean OR operators are ideal for including synonyms in your query when there are words that are relevant to your search, but you don't know whether they are actually present on the pages you are seeking.
NOT
The Boolean NOT is exactly the same as the - sign in search math, specifying what terms MUST NOT occur on the pages returned.
“ ”
Enclosing your query in quotes tells the search engine that it must only look for occurences of the EXACT phrase you've entered. | | | Once you've implemented your search strategy and have a good list of results, it's time to evaluate them.
Remember, all information is not equal. Anyone can find large quantities of information on the Internet. You want to assess the quality. Here are some things you may want to keep in mind:
Who created the site? By determining who created the site, you can:
- Assess potential bias in the content.
For instance, a site created by PETA (People for the Ethical Treatment of Animals) may have a much different perspective on pigs than a site created by the National Pork Producers Council.
- Rate the site's credibility.
An article on the US government's Food and Drug Administration web site, noting the side effects of certain drugs, may be more reliable than a similar article located on the drug company's site.
Top Of Page - Establish the agenda of the site's developers.
Is the purpose of the web site to inform, or to sell? Is it a personal home page, or is it a corporate site? EVERY site has an agenda. Web sites don't just pop into existence of their own volition. They are created by someone who has a reason for doing so.
For example:
Even “The Little Engine That Could” has an agenda. Although the mission of the “Little Engine” is to provide an innovative free service that allows users to participate, the underlying agenda lies in the hope that users will come to know and trust the site well enough to use the Shopping Center when making online purchases, thus assuring the service's continued existence.
- Who is the intended audience? It is usually easy to determine the targeted audience by looking at the vocabulary used in the article. Articles intended for academics and professionals tend to contain more complex language than those written for the general public.
- When was the site last updated? If you're gathering historical data, this may not be very important. However, if you are researching subjects where the most current information available is critical, you will prefer finding web pages that have been recently updated.
Top Of Page - Is the site easy to use? If extracting the information you want appears to be difficult, or you can't even find it once you're on the site, you might want to ask yourself how much time you're willing to spend trying to get information someone doesn't appear to be very interested in providing.
- Is the information relevant? There are articles written from nearly every point of view on every subject immaginable on the Internet. Do the articles you've found address the right aspects of the your topic?
| | | Using the above techniques, you should be able to locate authoritative material on any subject you choose. Although much of the information you will find is free for public use (i.e. “in the Public Domain”), there is also a lot of material that is copyrighted. Please respect intellectual property. Never claim someone elses work as your own, or worse, sell it for profit. Always give credit where credit is due.
If you intend to quote someone elses work verbatim, ask permission to do so. Individuals and self-publishers often copyright their materials more as a “by line” than for any other reason, and will be pleased when you show an interest in their work. Larger publishers and corporations take a more serious stance, but they usually have standard guidelines for the use of their materials. In any case, you may be surprised at how easy it is to get permission. Regardless of size, all publishers take a dim view on copyright infringement. They consider it theft, and they are correct.
If you will be writing for yourself, the choice of words is your own, and you intend to use someone else's work for research only, you aren't really required to give credit to others. However, if you intend to use exerpts (even paraphrased), be courteous enough to mention the original author as your source. This lends authority and credibility to your work, and recognizes your sources for their efforts. Remember, you would want nothing less for yourself if you're positions were reversed.
WHAT OUR LAWYERS MAKE US SAY: The above is not intended to be, nor should it be construed as, a treatise on copyright law. While believed to be accurate and current, the above is provided for informational purposes only. It is intended for the convenience of the users of this web site, and is not a legal opinion. Users wishing to obtain legal advice on copyright law should consult an experienced attorney.
| | | Barlow, Linda.
"The Spider's Apprentice: A Helpful Guide to Web Search Engines."
http://www.monash.com/spidap.html.
2001
Botos, Michael.
"Michael's Internet Finding Tips."
http://www.botos.com/train/searchit3a.html.
2000
Elkordy, Angela.
"Web Searching, Sleuthing, and Sifting."
http://www.thelearningsite.net/cyberlibrarian/searching/ismain.html.
2000
Habib, David P and Baillot, Robert L.
"How to Search the World Wide Web: A Tutorial for Beginners and Non-Experts."
http://204.17.98.73/midlib/tutor.htm.
2001
Liu, Jian.
"Guide to Meta-Search Engines."
http://www.indiana.edu/~librcsd/search/meta.html.
1999
LookSmart Ltd.,
"Search-Engine Guide."
http://www.looksmart.com/help/searchengs.html.
1999
Mamma
"The Mother of All Search Engines."
http://www.mamma.com/n_tips.html.
1999
Teoma
"Search with Authority: The Teoma Difference."
http://static.wc.teoma.com/docs/teoma/about/searchwithauthority.html.
2002
Tyner, Ross and Slaney, Walter.
"Sink or Swim: Internet Search Tools & Techniques."
http://www.sci.ouc.bc.ca/libr/connect96/search.htm#strategy.
2001
Yahoo
"Searching Yahoo!"
http://help.yahoo.com/help/us/ysearch/.
2002 | |
APS Home | Contact Us | eStore Demo | LETC Home
Copyright © 2002 - Allied Professional Services |
