A Novel Framework Using Zero Shot Learning Technique for a Non-Factoid Question Answering System

Autor/inn/en	R., Akila Devi T.; Sathick, K. Javubar; Khan, A. Abdul Azeez; Raj, L. Arun
Titel	A Novel Framework Using Zero Shot Learning Technique for a Non-Factoid Question Answering System
Quelle	In: International Journal of Web-Based Learning and Teaching Technologies, 16 (2021) 6, Artikel 12 (13 Seiten)Infoseite zur Zeitschrift PDF als Volltext Verfügbarkeit
Zusatzinformation	ORCID (Raj, L. Arun)
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
ISSN	1548-1093
Schlagwörter	Natural Language Processing; Artificial Intelligence; Classification; Responses; Information Processing + Suchen Sie Ihr Suchwort? Natürliche Sprache; Künstliche Intelligenz; Classification system; Klassifikation; Klassifikationssystem; Informationsverarbeitung
Abstract	Non-Factoid Question Answering (QA) is the next generation of textual QA systems, which gives passage level summaries for a natural language query, posted by the user. The main issue lies in the appropriateness of the generated summary. This paper proposes a framework for non-factoid QA system, which has three main components: (1) a deep neural network classifier, which produces sentence vector considering word correlation and context; (2) zero shot classifier that uses a multi-channel Convolutional Neural Network (CNN), to extract knowledge from multiple sources in the knowledge accumulator. This output acts as a knowledge enhancer that strengthens the passage level summary; and (3) summary generator that uses Maximal Marginal Relevance (MMR) algorithm, which computes similarity among the query related answer and the sentences from zero shot classifier. This model is applied on the datasets WikiPassageQA and ANTIQUE. The experimental analysis shows that this model gives comparatively better results for WikiPassageQA dataset. (As Provided).
Anmerkungen	IGI Global. 701 East Chocolate Avenue, Hershey, PA 17033. Tel: 866-342-6657; Tel: 717-533-8845; Fax: 717-533-8661; Fax: 717-533-7115; e-mail: journals@igi-global.com; Web site: https://www.igi-global.com/journals/
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01