Late last month, a significant code leak revealed a trove of data offering insights into the inner workings of Yandex, the Russian search engine company. This leak has provided an unprecedented look into the company’s sophisticated ranking algorithm.
Yandex revealed in a statement that the leak, which includes more than 1,500 lines of code, originated in the company’s open-source repository on GitHub. The leaked code reveals the company’s internal ranking algorithm, which is used to determine the placements of websites in search query results.
What’s of particular interest, is that the algorithm not only accounts for keywords, URLs and other standard factors usually associated with ranking, but also factors in complex metrics such as the navigational path of visitors, how quickly visitors load a web page, how much scrolling occurs on a website and even the “time and attention” a given visitor spends on the web page.
Yandex has stayed relatively tight-lipped about the leak, but it is clear from the code that the company is striving for accuracy and fairness when it comes to determining search results. As webmaster take into account this ranking algorithm to optimize their websites, they can be sure that they’ll be doing so with the same criteria that Yandex is using.
The leak has also revealed some of the other tactics Yandex uses to power its search engine. These tactics include using artificial intelligence to identify entities on websites and using machine learning to detect and filter out low-quality websites.
The Yandex code leak has revealed significant information about the company’s ranking algorithm, however, it is still unclear how much of a role external factors – such as news organizations or government entities – can have on search rankings.
As the code leak continues to be studied and as Yandex continues to refine its algorithm, hopefully some of the controversial factors used to power the Russian search engine will be revealed.
Hey Subscribe to our newsletter for more articles like this directly to your email.