Unlocking the Semantics of Roget's Thesaurus Using Formal Concept Analysis

摘要

Roget's Thesaurus is a semantic dictionary that is organized by concepts rather than words. It has an elaborate implicit structure that has not, in the 150 years since its inception, been made explicit. Formal Concept Analysis (FCA) is a tool that can be used by researchers for the organization, analysis and visualization of complex hidden structures. In this paper we illustrate two ways in which FCA is being used to explicate the implicit structures in Roget's Thesaurus: implications and Type-10 chain components.