Word2vec Based System for Recognizing Partial Textual Entailment

VÍTA, Martin and Vincent KRÍŽ. Word2vec Based System for Recognizing Partial Textual Entailment. In Ganzha, M Maciaszek, L Paprzycki, M. PROCEEDINGS OF THE 2016 FEDERATED CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS (FEDCSIS). NEW YORK: IEEE. p. 513-516. ISBN 978-83-60810-90-3. doi:10.15439/2016F419. 2016.

Basic information

• Original name: Word2vec Based System for Recognizing Partial Textual Entailment
• Authors: VÍTA, Martin (203 Czech Republic, belonging to the institution) and Vincent KRÍŽ (203 Czech Republic).
• Edition: NEW YORK, PROCEEDINGS OF THE 2016 FEDERATED CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS (FEDCSIS), p. 513-516, 4 pp. 2016.
• Publisher: IEEE

Other information

• Original language: English
• Type of outcome: Proceedings paper
• Field of Study: 10201 Computer sciences, information science, bioinformatics
• Confidentiality degree: is not subject to a state or trade secret
• Publication form: electronic version available online
• RIV identification code: RIV/00216224:14330/16:00094081
• Organization unit: Faculty of Informatics
• ISBN: 978-83-60810-90-3
• ISSN: 2300-5963
• Doi: http://dx.doi.org/10.15439/2016F419
• UT WoS: 000392436600073
• Keywords in English: textual entailment; word2vec model; faceted entailment
• Tags: firank_B
• Tags: International impact, Reviewed

Abstract

Recognizing textual entailment is typically considered as a binary decision task whether a text T entails a hypothesis H. Thus, in case of a negative answer, it is not possible to express that IT is "almost entailed" by T. Partial textual entailment provides one possible approach to this issue. This paper presents an attempt to use word2vec model for recognizing partial (faceted) textual entailment. The proposed approach does not rely on language dependent NIT tools and other linguistic resources, therefore it can he easily implemented in different language environments where word2vec models are available.