Database search engine: Difference between revisions

Revision as of 21:24, 18 June 2015

Revision as of 14:54, 19 June 2015

Line 1:

{{copyedit|date=June 2015}}

Categories of [[search engine]] software include: Webweb search or full-text search (i.e. [[Lucene]]), database or structured data search such as [[Diesel point]], and mixed or [[enterprise search]] like [[Google Search Appliance]]. The largest web internet directories, such as [[Google]] and [[Yahoo!]], utilize hundreds of thousands of computers to process billions of web pages, returning results for thousands of searches per second. High volumes of queries and text processing require the software to run in a highly distributed environment with a high degree of redundancy.

=== Modern search engine components: ===

Searching for text-based content in [[databases]] or [[structured data]] formats (such as [[XML]], [[Comma-separated values|CSV]], etc.) presents special challenges and opportunities, which a number of specialized search engines resolve. Databases are slow when solving complex queries, which have multiple logical or [[string matching]] arguments. Databases allow logical queries - for instance, the use of multi-field [[boolean logic]] - which full-text searches don't. "Crawling" (a human by-eye search) is not necessary to find information stored in a database, since the data is already structured, although it is often necessary to index the data in a more compact form that is designed to allow for faster searches.

Database search engines were initially (and still usually are) included with major database software products. As such, they are usually called "indexing engines". However, these indexing engines are relatively limited in their ability to customize indexing formats (such as compounding, normalization, transformation, [[transliteration]], etc.). Usually, they do not provide sophisticated data matching technology (such as [[string matching]], [[boolean logic]], algorithmic methods, search scripting, etc.).

In more advanced database search systems, relational databases are indexed by compounding multiple tables into a single table containing only the fields that need to be "queried" (displayed in search results). The actual data-matching engines can include any functions, such as basic string matching, normalization, and transformation. Database search technology is heavily used by many large entities, such as government database services, e-commerce companies, web advertising platforms, telecommunications service providers, etc.

==See also==