Developing A Site Search Engine With PHP And MySQL - Search Engine Theory
(Page 2 of 7 )
The search engine that we're going to create in this article is extremely basic, but still adheres to the theories of any other search engine: a user enters one or more keywords and a set of results are returned based on these keywords.
Search engines have been one of the reasons why the Internet has become such a popular medium for information storage and retrieval. For many of the modern search engines (such as Google and Yahoo), you can think of the way they work as giant yellow pages. All of their data is indexed in a way that a set of algorithms can easily be applied to it to return matching records. These records contain the title of the matching web site, as well as its ranking, short description, and URL.
Because these sites have millions (if not billions) of records indexed, the results normally span several pages, with the most relevant results being displayed first. A typical web user will not view past the third page of search results, so it is imperative that the first three pages contains links matching the users keywords as closely and as accurately as possible.
One of the other design requirements of any search engine is speed. Google has shown that a search engine doesn't have to include fancy listings or advanced options to become popular. In just a couple of years, Google has become one of the most popular (if not the most) search engines on the Internet. If you're after a good read, then checkout the
history of google. It's quite interesting.
We're going to try and make our search engine as fast as possible, limiting its functionality to non-duplicated, basic keyword recognition.
Next: Creating the database >>
More MySQL Articles
More By Mitchell Harper
/ 113 
