IP Library Granted Patent US 10,757,130
Granted Patent B2
US 10,757,130 · App. 16/167,855 · Granted Aug 25, 2020

Message authenticity and risk assessment

Inventors: Scot Free Kennedy (San Francisco, CA); Vidur Apparao (San Mateo, CA); Kevin Mandich (San Mateo, CA)
Assignee: Agari Data, Inc.
H04L63/1433H04L51/04H04L61/2007H04L63/1466H04L63/1483
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,757,130
App. No.
16/167,855
Granted
Aug 25, 2020
Kind
B2
Abstract

Assessing a risk of a message is disclosed. A sender specified by the message is identified. A measure of authenticity that the sender specified by the message is an actual sender of the message is determined using at least one sender model associated with the sender. The sender model was at least in part automatically generated using one or more previously observed messages. The measure of authenticity is utilized to perform a risk assessment of the message.

Claims (48)

1. A system, comprising:

a communication interface configured to receive an electronic message; and

one or more hardware processors coupled with the communication interface and configured to:

identify a sender specified by the electronic message based at least in part on an internet protocol (IP) address associated with the electronic message;

determine a measure of authenticity that the sender specified by the electronic message is an actual sender of the electronic message using at least one sender model associated with the sender, including by being configured to:

determine whether the IP address associated with the electronic message is within or adjacent to a block of IP addresses that are known or authorized to send messages for the sender;

determine a neighborhood component factor value based on the determination of whether the IP address associated with the electronic message is within or adjacent to the block of IP addresses that are known or authorized to send messages for the sender;

determine whether one or more attributes included in a body of the message match attributes historically seen in bodies of legitimate messages sent by the sender; and

determine a message attribute component factor value based on the determination of whether the one or more attributes included in the body of the message match attributes historically seen in the bodies of the legitimate messages sent by the sender;

utilize the measure of authenticity to determine a measure of risk associated with the electronic message, wherein the measure of authenticity is based at least in part on the neighborhood component factor value and the message attribute component factor value, wherein the measure of risk identifies a likelihood that the electronic message is an undesired message; and

deliver the electronic message to a recipient of the electronic message based on whether the determined measure of risk is above a delivery threshold.

2. The system of claim 1 , wherein the system is an email gateway server.

3. The system of claim 1 , wherein identifying the sender specified by the electronic message includes identifying a network domain of the sender of the electronic message.

4. The system of claim 1 , wherein the at least one sender model specifies message properties that are characteristic of authentic messages sent by or on behalf of the sender.

5. The system of claim 4 , wherein the measure of authenticity is based at least in part by comparing message properties of the electronic message to message properties specified in the at least one sender model.

6. The system of claim 1 , wherein the at least one sender model specifies one or more of the following determined to be associated with the sender: one or more IP addresses, one or more ranges/blocks of IP addresses, one or more network hostnames/domains, and one or more autonomous system numbers.

7. The system of claim 1 , wherein the at least one sender model was at least in part automatically generated using one or more previously received messages from the sender.

8. The system of claim 1 , wherein the processor is further configured to determine a measure of reputation of the sender.

9. The system of claim 8 , wherein the measure of reputation of the sender identifies a general likelihood that the sender would send a message that is of value to the recipient.

10. The system of claim 8 , wherein the measure of reputation of the sender is based in part on a local reputation of the sender, wherein the local reputation of the sender is specific to the recipient of the electronic message.

11. The system of claim 8 , wherein the measure of reputation of the sender is based at least in part on at least one of a similarity factor value that indicates whether a domain of the sender mimics a domain of another sender, a domain registration history of the sender, a Domain Name Service activity of the sender, or a statistic history of messages sent by the sender.

12. The system of claim 8 , wherein the measure of risk for the electronic message corresponds to the measure of authenticity and the measure of reputation of the sender.

13. The system of claim 1 , wherein the one or more attributes included in the body of the electronic message include one or more of words, phrases, links, and attachments.

14. The system of claim 1 , wherein the processor is further configured to determine whether to deliver the electronic message.

15. The system of claim 1 , wherein the processor is further configured to label the electronic message or report a risk associated with the electronic message.

16. The system of claim 1 , wherein the at least one sender model is configured to determine whether the IP address associated with the electronic message is associated with an IP address associated with message servers that are known or authorized to send messages for the sender.

17. The system of claim 1 , wherein the at least one sender model is configured to determine whether the IP address associated with the electronic message is associated with an autonomous system number associated with message servers that are known or authorized to send messages for the sender.

18. The system of claim 1 , wherein the at least one sender model is configured to determine whether the IP address associated with the electronic message is associated with a network hostname/domain associated with message servers that are known or authorized to send messages for the sender.

19. A method, comprising:

receiving an electronic message;

identifying a sender specified by the electronic message based at least in part on an internet protocol (IP) address associated with the electronic message;

determining a measure of authenticity that the sender specified by the electronic message is an actual sender of the electronic message using at least one sender model associated with the sender, including by:

determining whether the IP address associated with the electronic message is within or adjacent to a block of IP addresses that are known or authorized to send messages for the sender;

determining a neighborhood component factor value based on the determining of whether the IP address associated with the electronic message is within or adjacent to the block of IP addresses that are known or authorized to send messages for the sender;

determining whether one or more attributes included in a body of the message match attributes historically seen in bodies of legitimate messages sent by the sender; and

determining a message attribute component factor value based on the determining of whether the one or more attributes included in the body of the message match attributes historically seen in the bodies of the legitimate messages sent by the sender;

utilizing the measure of authenticity to determine a measure of risk associated with the electronic message, wherein the measure of authenticity is based at least in part on the neighborhood component factor value and the message attribute component factor value, wherein the measure of risk identifies a likelihood that the electronic message is an undesired message; and

delivering the electronic message to a recipient of the electronic message based on whether the determined measure of risk is above a delivery threshold.

20. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving an electronic message;

identifying a sender specified by the electronic message based at least in part on an internet protocol (IP) address associated with the electronic message;

determining a measure of authenticity that the sender specified by the electronic message is an actual sender of the electronic message using at least one sender model associated with the sender, including by:

determining whether the IP address associated with the electronic message is within or adjacent to a block of IP addresses that are known or authorized to send messages for the sender;

determining a neighborhood component factor value based on the determining of whether the IP address associated with the electronic message is within or adjacent to the block of IP addresses that are known or authorized to send messages for the sender;

determining whether one or more attributes included in a body of the message match attributes historically seen in bodies of legitimate messages sent by the sender; and

determining a message attribute component factor value based on the determining of whether the one or more attributes included in the body of the message match attributes historically seen in the bodies of the legitimate messages sent by the sender;

utilizing the measure of authenticity to determine a measure of risk associated with the electronic message, wherein the measure of authenticity is based at least in part on the neighborhood component factor value and the message attribute component factor value, wherein the measure of risk identifies a likelihood that the electronic message is an undesired message; and

delivering the electronic message to a recipient of the electronic message based on whether the determined measure of risk is above a delivery threshold.

Assignments (6)
TERMINATION AND RELEASE OF FIRST LIEN INTELLECTUAL PROPERTY SECURITY INTEREST RECORDED AT REEL/FRAME 57157/0206 Recorded Nov 24, 2025
From: JEFFERIES FINANCE LLC
To: AGARI DATA, INC.
Reel/Frame 073769/0945 →
TERMINATION AND RELEASE OF SECOND LIEN INTELLECTUAL PROPERTY SECURITY INTEREST RECORDED AT REEL/FRAME 57157/0265 Recorded Nov 21, 2025
From: ACQUIOM AGENCY SERVICES LLC
To: AGARI DATA, INC.
Reel/Frame 073662/0811 →
ASSIGNMENT OF INTELLECTUAL PROPERTY SECURITY AGREEMENT Recorded Aug 14, 2025
From: GOLUB CAPITAL MARKETS LLC (AS EXISTING AGENT)
To: ACQUIOM AGENCY SERVICES LLC (AS SUCCESSOR COLLATERAL AGENT)
Reel/Frame 072471/0665 →
FIRST LIEN INTELLECTUAL PROPERTY SECURITY AGREEMENT Recorded Aug 10, 2021
From: AGARI DATA, INC.
To: JEFFERIES FINANCE LLC, AS COLLATERAL AGENT
Reel/Frame 057157/0206 →
SECOND LIEN INTELLECTUAL PROPERTY SECURITY AGREEMENT Recorded Aug 10, 2021
From: AGARI DATA, INC.
To: GOLUB CAPITAL MARKETS LLC, AS COLLATERAL AGENT
Reel/Frame 057157/0265 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 22, 2021
From: KENNEDY, SCOT; APPARAO, VIDUR; MANDICH, KEVIN
To: AGARI DATA, INC.
Reel/Frame 056012/0936 →
Continuity (2)
Continuation 15040288 · Feb 10, 2016
Related Publication 20190058727A1 · Feb 21, 2019