Search Engines - Power tools to find Specific Information

The word "search engine" is frequently informally used to describe many types of sites on the web. Web search engines are actually databases that contain references to thousands of resources. Users are able to interact with the database, submitting queries that "ask" the database if it contains resources that match specific criteria.

A software program, (known as a "robot", "spider" or "crawler") reads web pages, follows links between pages and sites, and collects information for later use. A crawler may send out requests to thousands of pages at a time. When it reaches an active site, it then sends robots out to all the sites that the page is linked to. In this way 10 million sites can be indexed each day, but there are hundreds of millions of sites on the Web. All this information is combined in a huge database called a full-text index. This index records not only the words found, but the location of the words and the phrases the words appear in for every Web page that the robots visit.

A web search engine provides an interface between the user and the underlying database. The web search engine runs the search string against the database, returns a list of resources that match the criteria, and displays the results for the user.

Search engines deal with specific words and phrases, not categories or information organization. No active human intelligence is involved in the generation of the index, and no filtering or evaluation of the quality of the pages take place. Every page on the Web is regarded as being just as valuable as every other page. Because each search engine collects resources differently, or uses different mathematical formulas, or algariths, to to the same query typed into several search engines is likely to produce different results.