Maria Anisimova, programmer-researcher of the machine learning team at Mail.ru Mail, said on November 7 that mail services manage to minimize the receipt of unwanted letters by users thanks to the development of machine learning technologies – ML models.
In conversation with “Gazeta.Ru” she explained that it is a mathematical function that is created to solve problems or predict a certain event based on available data. A machine learning model uses algorithms to find and remember patterns in data.
As Anisimova noted, systems recognize about 100 million spam emails per day, despite new tricks. Machine learning algorithms analyze all incoming information in letters: the sender’s reputation, URL, features of pictures and text – then identify suspicious patterns and protect users from possible threats, the expert pointed out.
ML models are regularly retrained thanks to user clicks on the “This is spam” button. This is how systems learn to quickly respond to even the newest tricks of fraudsters. ML systems also take blocking measures if the sender does not allow you to unsubscribe from mailings.
In addition, according to Aisimova, the work of two more ML models in mail is important – Auto-Registration and Hacking Analysis. The advanced ML-system of Autoregistrations blocks about 130 thousand suspicious registrations per day. At the same time, Hacking Analysis systems online analyze actions in the user’s mailbox and identify hacks within a few seconds after receiving a threat signal.
The specialist emphasized that by hacking email, attackers can gain access to the user’s personal data, including passwords and bank accounts. As soon as the system sees such suspicious activity, it immediately blocks the mailbox in order to hide the user’s personal data. Anisimova recommended protecting her account using two-factor authentication to avoid hacking.
Earlier, on June 1, a general spam blocking system was launched in Russia in June. It was clarified that a feedback form has appeared on the websites of the Federal Antimonopoly Service and mobile operators MTS, Tele2 and Megafon, through which you can file a complaint about spam calls and SMS.
#expert #spoke #systems #minimizing #spam