The natural language processing (NLP) community has developed a variety of methods for extracting and disambiguating information from research publications. However, they usually focus only on standard research entities such as authors, affiliations, venues, references and keywords. We propose a novel approach, which combines NLP and semantic technologies for generating from the text of research publications an OWL ontology describing software technologies used or introduced by researchers, such as applications, systems, frameworks, programming languages, and formats. The method was tested on a sample of 300 publications in the Semantic Web field, yielding promising results.
展开▼
