Lightweight parsing of classifications into lightweight ontologies

摘要: Understanding metadata written in natural language is a premise to successful automated integration of large scale, language-rich, classifications such as the ones used digital libraries. We analyze labels within classification by exploring their syntactic structure, we then show how this structure can be detect patterns that processed lightweight parser with an average accuracy 96.82%. This allows for deeper understanding semantics, which improve almost 18% automatic translation into ontologies required semantic matching, search and algorithms.