Title: Knowledge representation in generative lexicon: exploiting qualia structure for compound noun interpretation

Authors: Vandana Dwivedi; Sanjukta Ghosh

Addresses: Indian Institute of Technology (BHU) Varanasi, Varanasi, 221005, India ' Indian Institute of Technology (BHU) Varanasi, Varanasi, 221005, India

Abstract: The paper is an attempt towards building a lexical knowledge representation model using generative lexicon framework especially exploiting the qualia information of the nouns. The interpretation of compound nouns is a challenging task in any natural language processing application. Some most frequently occurring terms that are found in popular and scholarly texts of ayurveda are compound nouns in nature. The compound used in this paper is rasaayana, 'a process of rejuvenation or a product of that process'. This term occurs both as a head word as the second element or a non-head first element of a compound noun. The GL model is able to represent the natural polysemy of the word either as a process or a result and explain the meanings of the compound words created with it based on the qualia structure representation and the principle of coercion.

Keywords: generative lexicon; lexical knowledge representation; qualia roles; coercion; rasaayana compounds.

DOI: 10.1504/IJKL.2024.139552

International Journal of Knowledge and Learning, 2024 Vol.17 No.4, pp.331 - 342

Received: 12 Mar 2022
Accepted: 05 Nov 2022
Published online: 04 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article