SYSTEM AND METHOD FOR IDENTIFYING FORGED EMAILS

Patent №

US 9,740,858

Granted

2017-08-22

Filed 2015

Owner

TREND MICRO INCORPORATED

Lab

AI components

3

ml · nlp · planning

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

14798852

Forged emails are detected by extracting email address parts of a sender email address. The email address parts include an account name, a subdomain, and a base domain of the sender email address. The mutation ratio of the email address parts relative to reference strings are calculated to determine similarity of the email address parts to the reference strings. The mutation ratios are compared to ratio thresholds to identify suspicious email addresses, and the results of identifying suspicious email addresses are correlated with other computer security information to identify forged emails.

Machine learningNatural languagePlanningG06F 21/552G06F 21/563G06F 21/566H04L 51/212H04L 51/48

AI classification

Planning1.00
Machine learning0.98
Natural language0.72
AI hardware0.11
Knowledge representation0.08
Evolutionary computation0.00
Vision0.00
Speech0.00

Ownership

TREND MICRO INCORPORATED

assignment · 361920903

Assignors

WU, CHENG-HAN, LU, SHIH-WEI

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC