Detecting question intention using a K-Nearest neighbor based approach

Research output: Chapter in Book/Report/Conference proceedingConference contribution

449 Downloads (Pure)

Abstract

The usage of question answering systems is increasing daily. People constantly use question answering systems in order to find the right answer for different kinds of information, but the abundance of available data has made the process of obtaining relevant information challenging in terms of processing and analyzing it. Many questions classification techniques have been proposed with the aim of helping in understanding the actual intent of the user’s question. In this research, we have categorized different question types through introducing question type syntactical patterns for detecting question intention. In addition, a k-nearest neighbor based approach has been developed for question classification. Experiments show that our approach has a good level of accuracy in identifying different question types.
Original languageEnglish
Title of host publicationArtificial Intelligence Applications and Innovations
Subtitle of host publicationAIAI 2018 IFIP WG 12.5 International Workshops, SEDSEAL, 5G-PINE, MHDW, and HEALTHIOT, Rhodes, Greece, May 25-27, 2018, Proceedings
EditorsLazaros Iliadis, Ilias Maglogiannis, Vassilis Plagianakos
PublisherSpringer
Pages101-111
Number of pages11
ISBN (Electronic)978-3-319-92016-0
ISBN (Print)978-3-319-92015-3
DOIs
Publication statusPublished - May 2018
Event14th International Conference on Artificial Intelligence Applications and Innovations - Rhodes, Greece
Duration: 25 May 201827 May 2018
http://easyconferences.eu/aiai2018/

Publication series

NameIFIP Advances in Information and Communication Technology
PublisherSpringer
Volume520
ISSN (Print)1868-4238
ISSN (Electronic)1868-422X

Conference

Conference14th International Conference on Artificial Intelligence Applications and Innovations
Abbreviated titleAIAI 2018
Country/TerritoryGreece
CityRhodes
Period25/05/1827/05/18
Internet address

Fingerprint

Dive into the research topics of 'Detecting question intention using a K-Nearest neighbor based approach'. Together they form a unique fingerprint.

Cite this