All news

Russian algorithm ranks among world’s most in-demand machine learning systems

According to the MarkTechPost report, CatBoost’s high accuracy and processing speed, along with its broad versatility, allowed it to be included among six non-US technologies that have succeeded in becoming global scientific standards

MOSCOW, December 15. /TASS/. The Russian machine learning algorithm CatBoost has emerged as one of the most widely used machine learning tools in both fundamental and applied science, according to a report by the US-based outlet MarkTechPost. Scientists in more than 50 countries use it as frequently as the most popular US-developed counterparts, Yandex’s press service reported.

"CatBoost is the only Russian technology included in the global report. It is used in every thirtieth scientific paper and stands on a par with solutions developed by Google, Microsoft, Intel, Amazon, and other major technology companies and research institutions. Scientists in more than 50 countries use the algorithm, confirming that Russian technologies enjoy global recognition," the statement said.

According to the release, this conclusion was reached by analysts at MarkTechPost after examining which machine learning algorithms and AI systems were used by leading scientists worldwide who published 5,000 research papers in the UK-based journal Nature, one of the most prestigious and highly ranked scientific publications.

Among the most popular solutions of this kind was the Russian algorithm CatBoost, first released as open source in 2017 and representing an advanced form of so-called gradient boosting. This term refers to a mathematical approach that enables the rapid identification of complex and non-obvious patterns in tabular data by gradually reducing errors in the outputs of a machine learning system.

According to the MarkTechPost report, CatBoost’s high accuracy and processing speed, along with its broad versatility, allowed it to be included among six non-US technologies that have succeeded in becoming global scientific standards.

The outlet’s analysis also showed that the Russian algorithm has enabled the development of new methods for diagnosing tumors, Alzheimer’s disease and preterm birth, has been used to create new approaches to forecasting water quality indices, and has demonstrated high effectiveness in detecting fake accounts on social media platforms. This underscores that the development by Russian scientists is a universal tool for addressing a wide range of scientific challenges.