Specialized Corpora Processing with Automatic Extraction Tools

This research describes a protocol for specialized corpus analysis using natural language processing (NLP) tools to define semantic hierarchies of verbs in the specialized domain of Volcanology. The experimental analysis was carried out with a domain-specific corpus in English of approximately 500,000 tokens composed of dissertations and scientific articles in the domain of Volcanology. The combination of semantic and syntactic analysis results in the verb macro structure that illustrates the evolution of the meaning from more general to more specific verbs.