Generating Domain Terminologies using Root- and Rule-Based Terms

Published

December 21, 2018

Author(s)

Talapady N. Bhat, John T. Elliott, Ursula R. Kattner, Carelyn E. Campbell, Eswaran Subrahmanian, Ram D. Sriram, Jacob Collard, Monarch Ira

Abstract

Motivated by the need for exible, intuitive, reusable, and normalized ter- minology for the semantic web, we present a general approach for generat- ing sets of such terminologies from nat- ural language documents. The terms that this approach generates are root- and rule-based terms, generated by a series of rules designed to be exible, to evolve, and, perhaps most impor- tant, to protect against ambiguity and reduce semantically similar but syn- tactically distinct phrases to a normal form. This approach combines several linguistic and computational methods that can be automated with the help of training sets to quickly and consis- tently extract normalized terms. We discuss how this can be extended as natural language technologies improve and how the strategy applies to com- mon use-cases such as search, docu- ment entry and archiving, and cura- tion

Citation

Journal of Washington Academy of Sciences

Volume

104

Pub Type

Journals

Download Paper

Local Download

Keywords

Root and rule-based terminology, Material Genome Initiative, Natural Language Processing, Dependency tree, Semantic Web, federated effort for terminology building, Standard Ontology, Metadata, Systematic building of metadata, Rules for metadata, Metadata standard

Citation

Bhat, T. , Elliott, J. , Kattner, U. , Campbell, C. , Subrahmanian, E. , Sriram, R. , Collard, J. and Ira, M. (2018), Generating Domain Terminologies using Root- and Rule-Based Terms, Journal of Washington Academy of Sciences, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=919688 (Accessed April 19, 2024)