How eBay Uses AI to Identify Brands in Search Queries
A system that uses deep learning to recognize brand names in search queries and automatically improve search results by adding relevant product terms.
Original patent title: “Deep hybrid neural network for named entity recognition”
A system that uses deep learning to recognize brand names in search queries and automatically improve search results by adding relevant product terms. Granted to eBay Inc in 2023 with 21 claims and 2 forward citations.
Key facts
Coverage
What does this patent actually cover?
This system improves search accuracy by teaching a computer to understand that certain words in a search query belong together as a single brand name. It first breaks down words into individual characters using a deep neural network to understand their structure, then combines this with pre-trained word knowledge. It uses a bidirectional long short-term memory (LSTM) to look at the context of the whole sentence, and finally applies conditional random fields to pick the most likely label for each word. For example, if a user searches for 'Nike running shoes', the system identifies 'Nike' as a brand and may automatically add terms like 'apparel' or 'gear' to the search to return better results.
The gap
What does this patent NOT cover?
- Does not cover general-purpose entity recognition that is not tied to a search query augmentation process.
- Does not cover systems that identify entities without using both character-level convolutional layers and bidirectional LSTMs.
- Does not cover search augmentation that does not rely on the specific output of a sequential conditional random field classifier.
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
What made this novel
The system combines character-level information (the shape and structure of the word) with word-level embeddings (the meaning of the word), allowing the model to recognize brand names it has never seen before based on their character patterns.
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
eBay search query processing
Automated product category tagging in e-commerce
Brand-aware query expansion for online marketplaces
Why it matters
The bigger picture
In e-commerce, search is the primary way users find products. If a search engine fails to understand that 'New Balance' is a brand rather than a description of a person's physical state, the user gets irrelevant results. This patent describes a specific technical pipeline for ensuring that large-scale search platforms can accurately parse brand-specific intent to drive sales.
Filed
August 31, 2017
Granted
February 28, 2023
Market context
Who's building on this
Companies in this space
eBay continues to refine its search and discovery algorithms using these types of hybrid neural architectures. Major e-commerce platforms like Amazon and Alibaba also employ similar deep learning pipelines to parse user intent and improve search relevance.
Market impact
This patent reflects the industry-wide shift toward deep learning for natural language processing in search. By automating the identification of brand entities, companies can reduce the need for manual keyword mapping and improve the conversion rates of search queries into actual product purchases.
Claim 1 — Plain English
What this patent covers
This system improves search accuracy by teaching a computer to understand that certain words in a search query belong together as a single brand name. It first breaks down words into individual characters using a deep neural network to understand their structure, then combines this with pre-trained word knowledge. It uses a bidirectional long short-term memory (LSTM) to look at the context of the whole sentence, and finally applies conditional random fields to pick the most likely label for each word. For example, if a user searches for 'Nike running shoes', the system identifies 'Nike' as a brand and may automatically add terms like 'apparel' or 'gear' to the search to return better results.
The clever bit
The system combines character-level information (the shape and structure of the word) with word-level embeddings (the meaning of the word), allowing the model to recognize brand names it has never seen before based on their character patterns.
What it does not cover
- Does not cover general-purpose entity recognition that is not tied to a search query augmentation process.
- Does not cover systems that identify entities without using both character-level convolutional layers and bidirectional LSTMs.
- Does not cover search augmentation that does not rely on the specific output of a sequential conditional random field classifier.
Patent timeline
Application submitted to the patent office
Application published, typically 18 months after filing
Patent officially issued
PatentBrief Score
Impact Score
Moderate
Citation count
10/40
Early citations
Claim breadth
14/20
Broad claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →
Recency
20/20
Granted within 5 years
Assignee scale
0/20
Independent or smaller assigneeassigneeThe entity that owns the patent — usually the inventor's employer or a company.Read more →
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$52K – $166K
Midpoint $104K · 11.2 yr remaining · industry ×1.6
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
The original legal language
Original claims
21 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Xin, Y., Ruvini, J., & Hart, E. J. (2023). How eBay Uses AI to Identify Brands in Search Queries (U.S. Patent No. 11,593,558). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/11593558/no-language-left-behind-nllb
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US11593558"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
More to explore
More in AI & Machine Learning
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11615208 · 2023 · Capital One Services LLC
How Cloud Systems Automatically Create and Train AI Data Models
US 10402750 · 2019 · Facebook Inc
How Facebook Uses Deep Learning to Predict What You Might Like
New to patents?
Common Questions
Frequently Asked Questions
What does How eBay Uses AI to Identify Brands in Search Queries cover?
A system that uses deep learning to recognize brand names in search queries and automatically improve search results by adding relevant product terms.
Who owns patent US 11593558?
eBay Inc owns this patent, granted in 2023.
When does this patent expire?
This patent is expected to expire on February 28, 2043, when the invention enters the public domain.
What is patent US 11593558 cited by?
This patent has been cited by 2 later patents that build on its ideas.
What problem does this patent solve?
In e-commerce, search is the primary way users find products. If a search engine fails to understand that 'New Balance' is a brand rather than a description of a person's physical state, the user gets irrelevant results. This patent describes a specific technical pipeline for ensuring that large-scale search platforms can accurately parse brand-specific intent to drive sales.
What does this patent NOT cover?
Does not cover general-purpose entity recognition that is not tied to a search query augmentation process.
Same assignee
More from eBay Inc
Patent monitoring