|SE4SEE is an innovative grid-enabled search engine, developed in Turkey to specifically target countries in South-East Europe.
Image courtesy of the Turkish National Grid Initiative TR-Grid
Search Engine for South-East Europe is a grid-enabled, personalized, on-demand, country-specific, category-based search engine. It specifically targets countries in South-East Europe and differs from a traditional search engines in both its design philosophy and functionality.
The main features of SE4SEE include:
While traditional search engines crawl the entire Web every so often, SE4SEE starts an individual crawl for each user query, searching original copies of pages in the web to ensure the most up-to-date versions are evaluated.
Traditional search engines crawl the Web continuously; SE4SEE only initiates a crawl when it receives a user query. Users can also control crawl conditions, such as the number of download pages or the duration of a crawl. This use is suited to long-term query evaluation, where the user has no time constraints and the Web is searched for a period of hours or days.
As well as keyword-based searching, SE4SEE supports category-based searches. In category-based searching, a user selects a category, and a previously trained text classifier categorizes pages downloaded by the crawler. At the completion of the crawl, SE4SEE presents only those pages relevant to the selected category. Currently, only category searches within Turkey are possible, as the categorization engine has only been trained for Turkey.
SE4SEE provides country-specific searches, performed based on the language of the page, the country domain of the page URL, or the geographical locality of the hosting site.
And why grid?
The resources available on a grid (computational power, storage capacity and network bandwidth) alleviate performance concerns usually associated with this type of Web.
SE4SEE is based on grid infrastructure established as part of the SEE-GRID project, which is funded by the European Community.
Want more detailed information? All the details and data are at Architecture of a grid-enabled Web search engine, Cambazoglu et. al., Information Processing & Management, Vol 43, pp. 609-623, 2007.