Turkish Coreference Resolution

Tugba Pamay; Gulsen Eryigit

doi:10.1109/INISTA.2018.8466293

Turkish Coreference Resolution

Tugba Pamay, Gulsen Eryigit

Department of Artificial Intelligence and Data Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Citations (Scopus)

Abstract

This paper presents the state-of-the-art results in Turkish coreference resolution (CR) which is a task of determining sets of mentions which identify the same real-world entity (e.g. a person, a place, a thing, an event). The proposed system uses support vector machines and solves the CR task with a mention-pair model that basically accepts mention couples and decides on whether they are coreferential with each other or not. The results are evaluated on Marmara Turkish Coreference Corpus by using well-known evaluation metrics (viz. MUC, B³, BLANC and LEA). The introduced approach obtains F1 scores of 90.68% (MUC), 86.89% (B³), 85.13% (BLANC) and 78.34% (LEA) yielding an improvement of 9.12, 16.06, 13.08 and 12.57 percentage points respectively over a recent baseline system on Turkish CR. The paper introduces the system setup (SVM parameters and negative sampling strategy) as well as the selected features and analyzes the impact of these features on the Turkish CR task.

Original language	English
Title of host publication	2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018
Editors	Plamen Angelov, Tulay Yildirim, Lazaros Iliadis, Yannis Manolopoulos
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538651506
DOIs	https://doi.org/10.1109/INISTA.2018.8466293
Publication status	Published - 14 Sept 2018
Event	2018 IEEE International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018 - Thessaloniki, Greece Duration: 3 Jul 2018 → 5 Jul 2018

Publication series

Name	2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018

Conference

Conference	2018 IEEE International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018
Country/Territory	Greece
City	Thessaloniki
Period	3/07/18 → 5/07/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

Coreference Resolution
Natural Language Processing
Turkish

Access to Document

10.1109/INISTA.2018.8466293

Cite this

Pamay, T., & Eryigit, G. (2018). Turkish Coreference Resolution. In P. Angelov, T. Yildirim, L. Iliadis, & Y. Manolopoulos (Eds.), 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018 Article 8466293 (2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INISTA.2018.8466293

Pamay, Tugba ; Eryigit, Gulsen. / Turkish Coreference Resolution. 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018. editor / Plamen Angelov ; Tulay Yildirim ; Lazaros Iliadis ; Yannis Manolopoulos. Institute of Electrical and Electronics Engineers Inc., 2018. (2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018).

@inproceedings{5409bed3cdd945728c45344cdfcbc127,

title = "Turkish Coreference Resolution",

abstract = "This paper presents the state-of-the-art results in Turkish coreference resolution (CR) which is a task of determining sets of mentions which identify the same real-world entity (e.g. a person, a place, a thing, an event). The proposed system uses support vector machines and solves the CR task with a mention-pair model that basically accepts mention couples and decides on whether they are coreferential with each other or not. The results are evaluated on Marmara Turkish Coreference Corpus by using well-known evaluation metrics (viz. MUC, B3, BLANC and LEA). The introduced approach obtains F1 scores of 90.68% (MUC), 86.89% (B3), 85.13% (BLANC) and 78.34% (LEA) yielding an improvement of 9.12, 16.06, 13.08 and 12.57 percentage points respectively over a recent baseline system on Turkish CR. The paper introduces the system setup (SVM parameters and negative sampling strategy) as well as the selected features and analyzes the impact of these features on the Turkish CR task.",

keywords = "Coreference Resolution, Natural Language Processing, Turkish",

author = "Tugba Pamay and Gulsen Eryigit",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018 ; Conference date: 03-07-2018 Through 05-07-2018",

year = "2018",

month = sep,

day = "14",

doi = "10.1109/INISTA.2018.8466293",

language = "English",

series = "2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Plamen Angelov and Tulay Yildirim and Lazaros Iliadis and Yannis Manolopoulos",

booktitle = "2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018",

address = "United States",

}

Pamay, T & Eryigit, G 2018, Turkish Coreference Resolution. in P Angelov, T Yildirim, L Iliadis & Y Manolopoulos (eds), 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018., 8466293, 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018, Thessaloniki, Greece, 3/07/18. https://doi.org/10.1109/INISTA.2018.8466293

Turkish Coreference Resolution. / Pamay, Tugba ; Eryigit, Gulsen.
2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018. ed. / Plamen Angelov; Tulay Yildirim; Lazaros Iliadis; Yannis Manolopoulos. Institute of Electrical and Electronics Engineers Inc., 2018. 8466293 (2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Turkish Coreference Resolution

AU - Pamay, Tugba

AU - Eryigit, Gulsen

PY - 2018/9/14

Y1 - 2018/9/14

N2 - This paper presents the state-of-the-art results in Turkish coreference resolution (CR) which is a task of determining sets of mentions which identify the same real-world entity (e.g. a person, a place, a thing, an event). The proposed system uses support vector machines and solves the CR task with a mention-pair model that basically accepts mention couples and decides on whether they are coreferential with each other or not. The results are evaluated on Marmara Turkish Coreference Corpus by using well-known evaluation metrics (viz. MUC, B3, BLANC and LEA). The introduced approach obtains F1 scores of 90.68% (MUC), 86.89% (B3), 85.13% (BLANC) and 78.34% (LEA) yielding an improvement of 9.12, 16.06, 13.08 and 12.57 percentage points respectively over a recent baseline system on Turkish CR. The paper introduces the system setup (SVM parameters and negative sampling strategy) as well as the selected features and analyzes the impact of these features on the Turkish CR task.

AB - This paper presents the state-of-the-art results in Turkish coreference resolution (CR) which is a task of determining sets of mentions which identify the same real-world entity (e.g. a person, a place, a thing, an event). The proposed system uses support vector machines and solves the CR task with a mention-pair model that basically accepts mention couples and decides on whether they are coreferential with each other or not. The results are evaluated on Marmara Turkish Coreference Corpus by using well-known evaluation metrics (viz. MUC, B3, BLANC and LEA). The introduced approach obtains F1 scores of 90.68% (MUC), 86.89% (B3), 85.13% (BLANC) and 78.34% (LEA) yielding an improvement of 9.12, 16.06, 13.08 and 12.57 percentage points respectively over a recent baseline system on Turkish CR. The paper introduces the system setup (SVM parameters and negative sampling strategy) as well as the selected features and analyzes the impact of these features on the Turkish CR task.

KW - Coreference Resolution

KW - Natural Language Processing

KW - Turkish

UR - http://www.scopus.com/inward/record.url?scp=85055570942&partnerID=8YFLogxK

U2 - 10.1109/INISTA.2018.8466293

DO - 10.1109/INISTA.2018.8466293

M3 - Conference contribution

AN - SCOPUS:85055570942

T3 - 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018

BT - 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018

A2 - Angelov, Plamen

A2 - Yildirim, Tulay

A2 - Iliadis, Lazaros

A2 - Manolopoulos, Yannis

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018

Y2 - 3 July 2018 through 5 July 2018

ER -

Pamay T , Eryigit G. Turkish Coreference Resolution. In Angelov P, Yildirim T, Iliadis L, Manolopoulos Y, editors, 2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8466293. (2018 IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications, INISTA 2018). doi: 10.1109/INISTA.2018.8466293

Turkish Coreference Resolution

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this