A natural language approach to information management: tracking scientific advances through the structure of words


Andrew Hippisley, Chara Karavasili

University of Surrey, Guildford, Surrey, UK, GU2 7XH, a.hippisley@surrey.ac.uk




In scientific texts specialist words, or terms, express conceptual knowledge. We show that by looking at the use of a term and its family of derivatives over time we can have a tangible picture of how an underlying concept has evolved in scientific advances. This is because the structure of a word encases a core idea and how that idea has been extended in a particular direction. This paper is an outline of a research programme with some preliminary results from an analysis the term nucleus in a nine million word corpus of nuclear physics articles written over thirty-six years, from 1969 to the present day, representing a body of specialist knowledge of recognized growth over time.


information management, natural language approach, word-structure, diachronic corpus, specialist texts, knowledge life-cycle

Language(s) English
Full Paper