Development of a Method for Automatic Extraction of Ontology Entity Names from Natural Language Texts

2024

Closed access Metadata (CSL-JSON)
DOI 10.1109/ubmk63289.2024.10773503
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Domain ontologies play a crucial role in organizing, sharing, and reusing domain-specific knowledge within software systems. However, developing a software ontology is a time-consuming and complex process. To create such an ontology, a large number of relevant publications must be analyzed. This task of enriching the ontology with data from these sources can be streamlined and accelerated through the use of lexical and syntactic patterns derived from ontological design. This paper presents a method for the automated construction of ontologies within a scientific domain, leveraging a system of heterogeneous ontological design patterns (ODPs). This system includes ODPs tailored for ontology developers and automatically generated lexical and syntactic patterns, which can be used to enrich ontologies with information extracted from natural language texts.