NEC Corporation Link Prediction Using Knowledge Graph Embeddings and NLP
Summary
NEC Corporation has filed patent application US20260112488A1 (Application No. 19352593) for an information processing apparatus and method that performs link prediction by combining knowledge graph embeddings with sentence embeddings from a trained natural language model. The system generates triples from a query, creates sentences from those triples, calculates scores using both NLP embeddings and knowledge graph embeddings, then aggregates the scores for prediction. The invention targets improved accuracy at lower computational cost for link prediction tasks. Filing date was October 8, 2025; publication date April 23, 2026.
About this source
GovPing monitors USPTO Patent Applications - Health Informatics (G16H) for new healthcare & life sciences regulatory changes. Every update since tracking began is archived, classified, and available as free RSS or email alerts — 146 changes logged to date.
What changed
This published patent application describes NEC Corporation's link prediction technology that combines two approaches: scoring triples using knowledge graph embeddings and scoring sentences derived from those triples using sentence embeddings from a trained natural language model. The system aggregates scores from both approaches to generate predictions. CPC classifications G16H 50/20 and G16C 20/70 indicate the technology's application in health informatics and AI. This is a published application, not a granted patent—the invention remains under examination.
Technology companies developing knowledge graph applications, recommendation systems, or network analysis tools should monitor this filing for potential future patent claims that could affect similar technology development. The dual-approach scoring methodology (combining structured graph embeddings with unstructured text embeddings) represents a specific claimed technique that may have implications for product development in related fields.
Archived snapshot
Apr 23, 2026GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.
INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD
Application US20260112488A1 Kind: A1 Apr 23, 2026
Assignee
NEC Corporation
Inventors
Kenichiro AKAGI
Abstract
An object is to provide a link prediction technology capable of performing highly accurate prediction at lower cost. An information processing apparatus includes an acquisition unit for acquiring a query, a first generation unit for generating one or a plurality of triples with reference to the query, a second generation unit for generating a sentence from each of the one or the plurality of triples, a first calculation unit for acquiring sentence embeddings obtained by using a trained natural language model for each of the generated sentences and calculating a score of each of the sentence embeddings, a second calculation unit for calculating a score of knowledge graph embedding of each of the one or the plurality of triples, and an aggregation unit for aggregating the score calculated by the first calculation unit and the score calculated by the second calculation unit.
CPC Classifications
G16H 50/20 G16C 20/70
Filing Date
2025-10-08
Application No.
19352593
Mentioned entities
Parties
Related changes
Get daily alerts for USPTO Patent Applications - Health Informatics (G16H)
Daily digest delivered to your inbox.
Free. Unsubscribe anytime.
Source
About this page
Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission
Source document text, dates, docket IDs, and authority are extracted directly from USPTO.
The summary, classification, recommended actions, deadlines, and penalty information are AI-generated from the original text and may contain errors. Always verify against the source document.
Classification
Who this affects
Taxonomy
Browse Categories
Get alerts for this source
We'll email you when USPTO Patent Applications - Health Informatics (G16H) publishes new changes.
Subscribed!
Optional. Filters your digest to exactly the updates that matter to you.