Search engine development is much harder than most people realize. If you want a search engine for your site that works as well as Google or AltaVista, your best bet is probably to buy it, and expect to pay quite a bit.
On the other hand, if you can’t afford to build or buy a good search engine, you may be better off without one. A search engine that gives poor results will not do your users any favors.
Some search engines can be obtained for free, and some search engine software is available in the public domain. While these tools can be quite effective, be sure that they provide the search features that are needed for your application.
Traditional database searches work by comparing search strings to each record in the database, and they return a match if the keyword exactly matches the record (or is a substring of the item in the record).
Full-text retrieval systems try to address several problems with this approach. They speed things up by indexing the information, and they use several rules of thumb to estimate the degree of relevance of each match, based primarily on how many times the keyword appears in the matching document.
Indexing creates a list of all terms in the whole set of documents, with references back to the documents. Thus, at the time of the search, the search term is looked up in the index and the matching documents are immediately found. When more than one search term is used, the set of documents is combined from the list provided in the index for each term.
No Comments so far ↓
There are no comments yet...Kick things off by filling out the form below.