Recognizing textual entailment in non-English text via automatic translation into English

Partha Pakray, Snehasis Neogi, Sivaji Bandyopadhyay, Alexander Gelbukh

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

3 Citations (Scopus)

Abstract

We show that a task that typically involves rather deep semantic processing of text-being recognizing textual entailment our case study-can be successfully solved without any tools at all specific for the language of the texts on which the task is performed. Instead, we automatically translate the text into English using a standard machine translation system, and then perform all linguistic processing, including syntactic and semantic levels, using only English language linguistic tools. In this case study we use Italian annotated data. Textual entailment is a relation between two texts. To detect it, we use various measures, which allow us to make entailment decision in the two-way classification task (yes / no). We set up various heuristics and measures for evaluating the entailment between two texts based on lexical relations. To make entailment judgments, the system applies named entity recognition module, chunking, part-of-speech tagging, n-grams, and text similarity modules to both texts, all those modules being for English and not for Italian. Rules have been developed to perform the two-way entailment classification. Our system makes entailment judgments basing on the entailment scores for the text pairs. The system was evaluated on Italian textual entailment data sets: we trained our system on Italian development datasets using the WEKA machine learning toolset and tested it on Italian test data sets. The accuracy of our system on the development corpus is 0.525 and on the test corpus is 0.66, which is a good result given that no Italian-specific linguistic information was used. © 2013 Springer-Verlag.
Original languageAmerican English
Title of host publicationRecognizing textual entailment in non-English text via automatic translation into English
Pages26-35
Number of pages22
ISBN (Electronic)9783642377976
DOIs
StatePublished - 10 Apr 2013
EventLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) -
Duration: 1 Jan 2014 → …

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7630 LNAI
ISSN (Print)0302-9743

Conference

ConferenceLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Period1/01/14 → …

Fingerprint

Linguistics
Semantics
Syntactics
Processing
Learning systems
Module
Named Entity Recognition
Machine Translation
N-gram
Text
Tagging
Machine Learning
Heuristics

Cite this

Pakray, P., Neogi, S., Bandyopadhyay, S., & Gelbukh, A. (2013). Recognizing textual entailment in non-English text via automatic translation into English. In Recognizing textual entailment in non-English text via automatic translation into English (pp. 26-35). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7630 LNAI). https://doi.org/10.1007/978-3-642-37798-3_3
Pakray, Partha ; Neogi, Snehasis ; Bandyopadhyay, Sivaji ; Gelbukh, Alexander. / Recognizing textual entailment in non-English text via automatic translation into English. Recognizing textual entailment in non-English text via automatic translation into English. 2013. pp. 26-35 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{4a7db6ed9ca749e9a0b0a7be5e8823e4,
title = "Recognizing textual entailment in non-English text via automatic translation into English",
abstract = "We show that a task that typically involves rather deep semantic processing of text-being recognizing textual entailment our case study-can be successfully solved without any tools at all specific for the language of the texts on which the task is performed. Instead, we automatically translate the text into English using a standard machine translation system, and then perform all linguistic processing, including syntactic and semantic levels, using only English language linguistic tools. In this case study we use Italian annotated data. Textual entailment is a relation between two texts. To detect it, we use various measures, which allow us to make entailment decision in the two-way classification task (yes / no). We set up various heuristics and measures for evaluating the entailment between two texts based on lexical relations. To make entailment judgments, the system applies named entity recognition module, chunking, part-of-speech tagging, n-grams, and text similarity modules to both texts, all those modules being for English and not for Italian. Rules have been developed to perform the two-way entailment classification. Our system makes entailment judgments basing on the entailment scores for the text pairs. The system was evaluated on Italian textual entailment data sets: we trained our system on Italian development datasets using the WEKA machine learning toolset and tested it on Italian test data sets. The accuracy of our system on the development corpus is 0.525 and on the test corpus is 0.66, which is a good result given that no Italian-specific linguistic information was used. {\circledC} 2013 Springer-Verlag.",
author = "Partha Pakray and Snehasis Neogi and Sivaji Bandyopadhyay and Alexander Gelbukh",
year = "2013",
month = "4",
day = "10",
doi = "10.1007/978-3-642-37798-3_3",
language = "American English",
isbn = "9783642377976",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "26--35",
booktitle = "Recognizing textual entailment in non-English text via automatic translation into English",

}

Pakray, P, Neogi, S, Bandyopadhyay, S & Gelbukh, A 2013, Recognizing textual entailment in non-English text via automatic translation into English. in Recognizing textual entailment in non-English text via automatic translation into English. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7630 LNAI, pp. 26-35, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 1/01/14. https://doi.org/10.1007/978-3-642-37798-3_3

Recognizing textual entailment in non-English text via automatic translation into English. / Pakray, Partha; Neogi, Snehasis; Bandyopadhyay, Sivaji; Gelbukh, Alexander.

Recognizing textual entailment in non-English text via automatic translation into English. 2013. p. 26-35 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7630 LNAI).

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

TY - GEN

T1 - Recognizing textual entailment in non-English text via automatic translation into English

AU - Pakray, Partha

AU - Neogi, Snehasis

AU - Bandyopadhyay, Sivaji

AU - Gelbukh, Alexander

PY - 2013/4/10

Y1 - 2013/4/10

N2 - We show that a task that typically involves rather deep semantic processing of text-being recognizing textual entailment our case study-can be successfully solved without any tools at all specific for the language of the texts on which the task is performed. Instead, we automatically translate the text into English using a standard machine translation system, and then perform all linguistic processing, including syntactic and semantic levels, using only English language linguistic tools. In this case study we use Italian annotated data. Textual entailment is a relation between two texts. To detect it, we use various measures, which allow us to make entailment decision in the two-way classification task (yes / no). We set up various heuristics and measures for evaluating the entailment between two texts based on lexical relations. To make entailment judgments, the system applies named entity recognition module, chunking, part-of-speech tagging, n-grams, and text similarity modules to both texts, all those modules being for English and not for Italian. Rules have been developed to perform the two-way entailment classification. Our system makes entailment judgments basing on the entailment scores for the text pairs. The system was evaluated on Italian textual entailment data sets: we trained our system on Italian development datasets using the WEKA machine learning toolset and tested it on Italian test data sets. The accuracy of our system on the development corpus is 0.525 and on the test corpus is 0.66, which is a good result given that no Italian-specific linguistic information was used. © 2013 Springer-Verlag.

AB - We show that a task that typically involves rather deep semantic processing of text-being recognizing textual entailment our case study-can be successfully solved without any tools at all specific for the language of the texts on which the task is performed. Instead, we automatically translate the text into English using a standard machine translation system, and then perform all linguistic processing, including syntactic and semantic levels, using only English language linguistic tools. In this case study we use Italian annotated data. Textual entailment is a relation between two texts. To detect it, we use various measures, which allow us to make entailment decision in the two-way classification task (yes / no). We set up various heuristics and measures for evaluating the entailment between two texts based on lexical relations. To make entailment judgments, the system applies named entity recognition module, chunking, part-of-speech tagging, n-grams, and text similarity modules to both texts, all those modules being for English and not for Italian. Rules have been developed to perform the two-way entailment classification. Our system makes entailment judgments basing on the entailment scores for the text pairs. The system was evaluated on Italian textual entailment data sets: we trained our system on Italian development datasets using the WEKA machine learning toolset and tested it on Italian test data sets. The accuracy of our system on the development corpus is 0.525 and on the test corpus is 0.66, which is a good result given that no Italian-specific linguistic information was used. © 2013 Springer-Verlag.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84875871109&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=84875871109&origin=inward

U2 - 10.1007/978-3-642-37798-3_3

DO - 10.1007/978-3-642-37798-3_3

M3 - Conference contribution

SN - 9783642377976

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 26

EP - 35

BT - Recognizing textual entailment in non-English text via automatic translation into English

ER -

Pakray P, Neogi S, Bandyopadhyay S, Gelbukh A. Recognizing textual entailment in non-English text via automatic translation into English. In Recognizing textual entailment in non-English text via automatic translation into English. 2013. p. 26-35. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-37798-3_3