Yandex today released a new version of search on www.yandex.ru; the testing period on the engine lasted a month. The final version takes into account feedback from thousands of users of the search engine.

The new updated version of Yandex’s search program, named “Magadan”, has many improvements. The number of ranking factors has doubled; the new search technology can now quickly rank the most relevant web pages first, without making the user wait until it finishes scanning all pages.

Yandex.Search can now understand abbreviations and transliterated words. For example, the improved algorithm will show relevant results for the [IE] query, including those pages, which do not contain the actual abbreviation, but have the words “Internet Explorer” instead.

The [ford] search query retrieves web pages that have this word both in Latin and in Cyrillic. The renewed system can also process queries, which have some additional characters, such as in [Europa+] or [C#].

Yandex.Search can find the right websites even if the user’s query had typos. Every day the new Yandex search detects typos in 2.5 m user queries and offers correctly spelled variants. One million typos get corrected automatically.

Yandex’s new search index was expanded to include over a billion pages of non-Russian web resources.

The next version of Yandex’s search engine will have the name of “Nakhodka”, which continues the tradition of giving names of Russian cities to new versions of the search program.

Source: HitSearch

This entry was posted on Saturday, May 17th, 2008 at 5:54 am and is filed under Global, News. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.