When you give a search engine a keyword, how does it know which of the millions of pages in its index to sort through? It follows a set of rules known as search algorithms. As mentioned before, each particular search engine closely guards the exact mathematical formula it uses as a trade secret. Most search engines do follow some general rules.
One of the key ranking rules is using the location and frequency of keywords on a web page. So, pages with the search words in the HTML title tag are assumed to be more relevant to the topic than other pages. In effect, title tags are like a headline of a news story. Another major factor in how search engines determine relevancy is how keywords appear in relation to other words on the page. The higher a keyword's frequency, the more relevant the page is viewed. All the major search engines follow that rule.
Nobody does it the same, which is why you get different results when you do a search, even the exact same search on two different search engines. Some search engines like Google put extra emphasis on what sites are linked to the primary site ("link analysis"), while others, like Teoma, rank searches for "hubs" and "authorities" related to your query terms that form a "community" about the topic.
Search engines have found some creative ways to deal with bad queries. Certain phrases become very popular and are repeatedly submitted to search engines, so most search engine companies have developed pre-programmed results. For example, the 2002 movie Black Hawk Down is based on the true story of American soldiers in the battle of Mogadishu trying to capture Somalian warlords. So several search engine companies have developed standard responses to queries about the movie, directing you to that movie's website, details about the true events, bios of the stars, details on the lives of the real people involved in the real incident.
Search engines differ in the way they work. Say you want more information about chocolateflavored peanut butter. One search engine or subject directory might search its database first for "chocolate." Then it takes those results and searches them for the additional word "peanut." Then it takes those results and scans for the word "butter" then presents you with the results.
Another search tool might require you to submit a term using a plus sign (+), so your search term would be "chocolate+peanut+butter." Another might require you to submit "chocolate AND peanut AND butter." A fourth search engine might conduct three simultaneous searches - one for each word - and extract only the documents that the three sets have in common. Other search tools might use a combination of these approaches.
All of this is not very important until you are clear about what you want. For now, it is enough to understand that every one of these search engines works differently. Consider them different tools for different purposes. For example, AltaVista is especially good for finding international sites and information in languages other than English because the AltaVista search site has a built-in translation component. Google and AlltheWeb allow you to opt for image-only and graphics-only hits. Teoma and WiseNut customize your results into folders - essentially an automatic grouping feature that makes for easier searching.
Search engines also vary in their presentation of results. Some search engines list their results alphabetically, others by relevance ranking, and some use a combination of both criteria, or different criteria altogether. Search engines are programmed to rank results according to many factors, including the location and frequency of keywords on a page. For example, depending on the search engine, pages with keywords appearing in the title are assumed to be more relevant, and pages with keywords located higher up on a web page - in a headline or the first few paragraphs of text - are ranked higher than pages where the keywords appear lower on the page. Also, the more frequently a keyword is mentioned, the more relevant a web page is considered to be. The link search feature of some search engines can serve as a quick credibility check because the links identify other sites that consider this site to be credible.
Our website is not responsible for the information contained by this article. Webworldarticles.com is a free articles resource thus practically any visitor can submit an article. However if you notice any copyrighted material, please contact us and we will remove the article(s) in discussion right away.
This article was sent to us by:
Damian Lissle at
08272010
1. Natural language search engines and when to use them
All articles in this directory are property of their respective authors. Additionally, read our Privacy Policy
© 2010 WebWorldarticles.com - All Rights Reserved. Partners: Gunblade Saga