MOSCOW, August 22. /TASS/. Russian internet company Yandex launched an upgraded search platform version based on comparison of a query intent and a web page, the company said on Tuesday.
The new search engine version operates using "Korolyov" algorithm identifying exact needs of a user with the aid of the neural network, Yandex said. The neural network analyzes an Internet page in its entirety and not merely the headline. Yandex at the same time determines the page essence in advance at the indexing stage, the company said.
The other specific feature of the "Korolyov" algorithm is that it addresses the intent of other queries made by users to land on the page, Yandex said.
"Yandex’s position as the largest search engine in Russia creates a positive feedback loop for our deep neural network algorithm, which leads to superior search results for our users," the company said in its blog.
Yandex is the largest search platform in Russia. Its share of the Russian search market (including searches on mobile devices) totaled 54.3% on average in the second quarter of this year, according to Yandex.Radar analytical service data.