Towards building a corpus of Turkish referring expressions

Date

2012-05-31

Author

Acartürk, Cengiz
Çakır, Murat Perit

Metadata

Show full item record

Item Usage Stats

66
views

0
downloads

In this paper we report on the preliminary findings of our ongoing study on Turkish referring expressions used in situated dialogs. Situated dialogs of pairs of Turkish speakers were collected while they were engaged with a collaborative Tangram puzzle solving task, which was designed by Spanger et al (2011) in an effort to build a corpus of referring expressions in Japanese and English. The paper provides our preliminary results on the Turkish corpus and compares them with the findings of comparable studies conducted on Japanese and English referring expressions.

URI

http://www.lrec-conf.org/proceedings/lrec2012/workshops/02.Turkic%20Languages%20Proceedings.pdf
https://hdl.handle.net/11511/75318

Conference Name

LREC 2012 Language and Resources Evaluation Conference, (21 Mayıs 2012)

Collections

Graduate School of Informatics, Conference / Seminar

Suggestions

OpenMETU
Core

RELIABILITY AND VALIDITY OF THE TURKISH VERSION OF THE AUTOMATIC THOUGHTS QUESTIONNAIRE SAHIN, NH; SAHIN, N (Wiley, 1992-05-01) The present study investigated the reliability and validity of the ATQ in the Turkish cultural context. It is hoped that with this type of cross-cultural information, the universality of the proposed cognitive constructs can be advocated more reliably. The subjects were 345 university students; the split-half reliability of the ATQ was found to be r = .91 and Cronbach's alpha was .93; and item/total correlations ranged between .36 and .69. The Turkish ATQ correlated well with the BDI (r = .75), and it discr...
Transfer Learning Using Twitter Data for Improving Sentiment Classification of Turkish Political News Kaya, Mesut; Fidan, Guven; Toroslu, İsmail Hakkı (2013-10-29) In this paper, we aim to determine the overall sentiment classification of Turkish political columns. That is, our goal is to determine whether the whole document has positive or negative opinion regardless of its subject. In order to enhance the performance of the classification, transfer learning is applied from unlabeled Twitter data to labeled political columns. A variation of self-taught learning has been proposed, and implemented for the classification. Different machine learning techniques, including...
On Developing New Text and Audio Corpora and Speech Recognition Tools for the Turkish Language Salor Durna, Özgül; Pellom, Brian; Çiloğlu, Tolga; Hacıoğlu, Kadri; Demirekler, Mübeccel (null; 2002-09-16) This paper describes recent work towards development of new corpora and tools for Turkish speech research. This effort represents an on-going collaboration between the Center for Spoken Language Research (CSLR) at the University of Colorado and the Department of Electrical Engineering at the Middle East Technical University (METU). A new text corpus developed from Turkish newspapers’ text is described. In addition, a 193-speaker audio corpus and pronunciation lexicon for the Turkish language is developed. ...
Online Self-regulation Questionnaire: Validity and Reliability Study of Turkish Translation KİLİS, SELCAN; Yıldırım, Zahide (2018-04-01) This study aims to translate online self-regulated learning questionnaire (OLSQ) into Turkish and administer its validity and reliability issues. In the questionnaire, 24 items were translated into Turkish separately by nine experts, and then back-translation was conducted by two language experts. Pilot test was conducted with 321 students enrolled in fully online associate degree programs in a well-known public university in Adana, Turkey. In terms of validity issue, confirmatory factor analysis indicated ...
Backchannels in spoken Turkish Aytaç Demirçivi, Kadriye; Işık Güler, Hale; Department of English Language Teaching (2021-4-01) This study aims to identify all the non-lexical and lexical backchannels and different functions carried out by these backchannels in the Spoken Turkish Corpus. It also aims to investigate differences in the use of backchannels in naturally formed groups in the data. In order to achieve these aims, Spoken Turkish Corpus was used as the data source and EXMaRALDA tools were used to annotate functions of the backchannels. A sub-corpus was formed consisting of 61 conversations from three main settings:conversat...

Citation Formats

C. Acartürk and M. P. Çakır, “Towards building a corpus of Turkish referring expressions,” presented at the LREC 2012 Language and Resources Evaluation Conference, (21 Mayıs 2012), İstanbul, Türkiye, 2012, Accessed: 00, 2021. [Online]. Available: http://www.lrec-conf.org/proceedings/lrec2012/workshops/02.Turkic%20Languages%20Proceedings.pdf.