Goran Glavas

Title: Goran Glavas: Knowledge Graphs and Textual Fact Disambiguation

Introduction

Goran Glavas is an inventor associated with Wuerzburg, Germany, and NEC Corporation. His work includes methods for organizing information extracted from text and resolving ambiguity by connecting textual facts to knowledge graphs.

Latest Patents

Glavas is associated with 1 patent, titled “Grouping and Linking Facts from Text to Remove Ambiguity Using Knowledge Graphs.” The invention describes a method for disambiguating textual facts by linking them to a reference knowledge graph and adding entities and relations where needed. It ranks facts represented as subject-relation-object triples, then determines whether each element should connect to an existing knowledge-graph entry or become a new entry. The resulting enriched knowledge graph can support decision-making in areas such as drug development, material synthesis, inventory management, suspect detection from textual messages, and healthcare. The method may also use a machine-learning ranking module.

Career Highlights

Glavas’s patent work focuses on combining information extraction, machine learning, and knowledge graphs to make textual data more useful and less ambiguous. The described approach is designed to support applications across several technical and practical fields.

Collaborations

The patent information names Kiril Gashteovski and Gorjan Radevski among Glavas’s coworkers. The invention is associated with NEC Corporation.

Conclusion

Goran Glavas’s patent presents a knowledge-graph approach to clarifying facts extracted from text, with potential applications ranging from scientific research to healthcare and information management.