Patents by Inventor Petr A. Salnikov

Petr A. Salnikov has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11539726
    Abstract: Disclosed herein are systems and methods for generating heuristic rules for identifying spam emails based on fields in headers of emails. In one aspect, an exemplary method comprises, collecting statistical data on contents of a plurality of emails; analyzing the statistical data to identify different types of content, including headers or hyperlinks in said emails; grouping the emails into clusters based on types of content identified in said emails, wherein at least one cluster group being based on fields in headers of said emails; generating a hash from the most frequent combination of group of data in each cluster; formulating regular expressions based on analysis of hyperlinks of emails corresponding to the generated hashes; and generating heuristic rule for identifying spam emails by combining the hashes and the corresponding regular expressions, wherein the hash is generated based on fields in the headers of said emails.
    Type: Grant
    Filed: September 30, 2021
    Date of Patent: December 27, 2022
    Assignee: AO Kaspersky Lab
    Inventors: Roman A. Dedenok, Dmitry S. Golubev, Petr A. Salnikov
  • Publication number: 20220021692
    Abstract: Disclosed herein are systems and methods for generating heuristic rules for identifying spam emails based on fields in headers of emails. In one aspect, an exemplary method comprises, collecting statistical data on contents of a plurality of emails; analyzing the statistical data to identify different types of content, including headers or hyperlinks in said emails; grouping the emails into clusters based on types of content identified in said emails, wherein at least one cluster group being based on fields in headers of said emails; generating a hash from the most frequent combination of group of data in each cluster; formulating regular expressions based on analysis of hyperlinks of emails corresponding to the generated hashes; and generating heuristic rule for identifying spam emails by combining the hashes and the corresponding regular expressions, wherein the hash is generated based on fields in the headers of said emails.
    Type: Application
    Filed: September 30, 2021
    Publication date: January 20, 2022
    Inventors: Roman A Dedenok, Dmitry S. Golubev, Petr A. Salnikov
  • Patent number: 11165792
    Abstract: Disclosed herein are systems and methods for generating heuristic rules for identifying spam emails. In one aspect, an exemplary method comprises, collecting and analyzing statistical data on contents of a emails to identify different types of content, including headers or hyperlinks, grouping the emails into clusters based on identified types of content, at least one cluster including groups of fields in the headers of said emails, selecting at least one most frequent combination of groups of data in each cluster, generating a hash from the at least one most frequent combination of groups, formulating at least one regular expression based on an analysis of hyperlinks corresponding to the generated hashes, and generating at least one heuristic rule for identifying emails containing spam by combining at least one hash and the corresponding regular expression, wherein at least one hash is from sequences of fields in the headers of said emails.
    Type: Grant
    Filed: July 17, 2019
    Date of Patent: November 2, 2021
    Assignee: AO KASPERSKY LAB
    Inventors: Roman A. Dedenok, Dmitry S. Golubev, Petr A. Salnikov
  • Publication number: 20200314120
    Abstract: Disclosed herein are systems and methods for generating heuristic rules for identifying spam emails. In one aspect, an exemplary method comprises, collecting and analyzing statistical data on contents of a emails to identify different types of content, including headers or hyperlinks, grouping the emails into clusters based on identified types of content, at least one cluster including groups of fields in the headers of said emails, selecting at least one most frequent combination of groups of data in each cluster, generating a hash from the at least one most frequent combination of groups, formulating at least one regular expression based on an analysis of hyperlinks corresponding to the generated hashes, and generating at least one heuristic rule for identifying emails containing spam by combining at least one hash and the corresponding regular expression, wherein at least one hash is from sequences of fields in the headers of said emails.
    Type: Application
    Filed: July 17, 2019
    Publication date: October 1, 2020
    Inventors: Roman A. Dedenok, Dmitry S. Golubev, Petr A. Salnikov