IP Library › Granted Patent US 12,393,570
Granted Patent B2
US 12,393,570 · App. 18/437,649 · Granted Aug 19, 2025

Method and system for address verification

Inventors: Marcus Gartner (San Francisco, CA); David Currie (San Francisco, CA)
Assignee: Lob.com, Inc.
G06F16/2365G06F16/90344G06N20/00
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 12,393,570
App. No.
18/437,649
Granted
Aug 19, 2025
Kind
B2
Abstract

The method for address verification preferably includes: receiving an unverified address; parsing the unverified address into address elements; determining a candidate address set based on the address elements; determining an address comparison set from the verified address database; selecting an intended address from the address comparison set; optionally facilitating use of the intended address; and optionally determining and providing a call to action based on the intended address.

Claims (38)

1. A method, comprising:

receiving an unverified address;

generating a candidate address set based on the unverified address;

determining an address comparison set from an address database based on the candidate address set using a hierarchical search, comprising:

determining whether a zip code for a candidate address of the candidate address set is within a set of zip codes associated with a city for the candidate address; and

adding a set of addresses associated with the zip code from the address database to the address comparison set;

selecting an intended address based on the address comparison set and the unverified address; and

determining a deliverability class for at least one of the unverified address or the intended address.

2. The method of claim 1 , wherein the deliverability class comprises at least one of deliverable, deliverable but missing one or more address element values, or undeliverable.

3. The method of claim 1 , wherein the deliverability class is determined based on a string distance between the unverified address and the intended address.

4. The method of claim 1 , wherein the deliverability class is determined using a machine learning model.

5. The method of claim 1 , wherein the deliverability class is provided to a sender endpoint using an API.

6. The method of claim 1 , wherein determining the deliverability class for the unverified address comprises:

for each candidate address of the candidate address set:

determining a candidate address from the candidate address set as undeliverable when the candidate address is not associated with an address of the address database; and

removing the candidate address from at least one of the candidate address set or not adding the candidate address to the address comparison set when the candidate address is undeliverable; and

determining the unverified address as undeliverable when at least one of: no candidate addresses of the candidate address set remain or no addresses from the address comparison set remain.

7. The method of claim 6 , wherein a notification is automatically sent to a user of the undeliverable unverified address.

8. The method of claim 1 , wherein the candidate address set comprises undeliverable addresses.

9. The method of claim 1 , further comprising determining a candidate deliverability class for a candidate address of the candidate address set, wherein the address comparison set is determined based on the candidate deliverability class.

10. The method of claim 1 , further comprising parsing the unverified address into a set of address elements, wherein the candidate address set is generated based on the set of address elements.

11. A system, comprising:

a processing system configured to:

receive an unverified address;

generate a candidate address set based on the unverified address;

determine an address comparison set from an address database based on the candidate address set, comprising:

when a zip code for a candidate address of the candidate address set is within a set of zip codes associated with a city for the candidate address, adding a set of addresses associated with the zip code from the address database to the address comparison set;

select an intended address based on the address comparison set and the unverified address; and

determine a deliverability confidence score for the intended address.

12. The system of claim 11 , wherein the deliverability confidence score for the intended address is determined based on a string distance of the intended address to the unverified address.

13. The system of claim 11 , wherein the deliverability confidence score for the intended address is determined based on addresses in the address database.

14. The system of claim 11 , wherein the deliverability confidence score for the intended address is determined based on historical delivery data associated with the intended address.

15. The system of claim 11 , wherein the deliverability confidence score is determined using a machine learning model.

16. The system of claim 11 , wherein the deliverability confidence score is provided to a user on an interface.

17. The system of claim 11 , wherein the intended address is selected based on a similarity score between an address from the address comparison set and the unverified address.

18. The system of claim 11 , wherein a physical asset is delivered to the intended address.

19. The system of claim 11 , wherein the unverified address is received as part of a batch of addresses in a single action, wherein the batch of addresses is processed by the system in parallel.

20. The system of claim 11 , wherein generating the candidate address set based on the unverified address comprises converting an existing address element of the unverified address into a phonetic encoding using a phonetic algorithm, wherein a candidate address of the candidate address set is generated by replacing the existing address element with a new address element associated with the phonetic encoding.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 22, 2024
From: GARTNER, MARCUS; CURRIE, DAVID
To: LOB.COM, INC.
Reel/Frame 066533/0886 →
Continuity (4)
Continuation 17514182 · Oct 29, 2021
Continuation In Part 17127906 · Dec 18, 2020
Provisional Application 62951327 · Dec 20, 2019
Related Publication 20240184768A1 · Jun 6, 2024
References Cited (24)
US 8341520B2 · Iakobashvili et al. · 2012 [cited by applicant]
US 8996523B1 · Fisher · 2015 [cited by examiner]
US 9977633B1 · Khafizova · 2018 [cited by applicant]
US 10819849B1 · Bondareva et al. · 2020 [cited by applicant]
US 11055355B1 · Monti et al. · 2021 [cited by applicant]
US 11188782B2 · Gartner et al. · 2021 [cited by applicant]
US 20030218770A1 · Field · 2003 [cited by applicant]
US 20040207867A1 · Stringham · 2004 [cited by applicant]
US 20090055206A1 · Orbke · 2009 [cited by applicant]
US 20090157733A1 · Kim et al. · 2009 [cited by applicant]
US 20090187733A1 · El-Ghazawi · 2009 [cited by applicant]
US 20110307476A1 · Sharma · 2011 [cited by examiner]
US 20130054491A1 · Chatow et al. · 2013 [cited by applicant]
US 20140079428A1 · Park et al. · 2014 [cited by applicant]
US 20140149309A1 · Park · 2014 [cited by examiner]
US 20160041895A1 · Galvin · 2016 [cited by examiner]
US 20160300250A1 · Rai et al. · 2016 [cited by applicant]
US 20160352976A1 · Kuroiwa · 2016 [cited by applicant]
US 20170374093A1 · Dhar · 2017 [cited by examiner]
US 20200278987A1 · Liu et al. · 2020 [cited by applicant]
JP 2013206235A · 2013 [cited by applicant]
“How Do We Route Print Jobs?”, Smart Order Routing, https://www.cloudprinter.com/order-routing, Nov. 9, 2020, 9 pages. [cited by applicant]
Chen, Anthony , “How We Solve Problems at Lob—Intelligent Mail Through Routing”, Lob Blog, https://www.lob.com/blog/how-we-solve-problems-at-lob-intelligent-mail-through-routing, Jul. 18, 2018, 15 pages. [cited by applicant]
Comber, Sam , “Machine learning innovations in address matching: A practical comparison of word2vec and CRFs”, wileyonlinelibrary.com/journal/tgis, Transactions in GIS, 2019. [cited by applicant]