Cross-lingual Transfer Learning for Javanese Dependency Parsing

Fadli Aulawi Al Ghiffari; Ika Alfina; Kurniawati Azizah

doi:10.18653/v1/2023.ijcnlp-srw.1

Cross-lingual Transfer Learning for Javanese Dependency Parsing

Fadli Aulawi Al Ghiffari, Ika Alfina, Kurniawati Azizah

Department of Computer Science

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

Abstract

While structure learning achieves remarkable performance in high-resource languages, the situation differs for under-represented languages due to the scarcity of annotated data.This study focuses on assessing the efficacyof transfer learning in enhancing dependency parsing for Javanese—a language spoken by 80million individuals but characterized by limited representation in natural language processing. We utilized the Universal Dependencies dataset consisting of dependency treebanks from more than 100 languages, including Javanese. We propose two learning strategies to train the model: transfer learning (TL) and hierarchical transfer learning (HTL). While TL onlyuses a source language to pre-train the model, the HTL method uses a source language and an intermediate language in the learning process. The results show that our best model usesthe HTL method, which improves performance with an increase of 10 % for both UAS and LAS evaluations compared to the baseline model

Original language	English
Title of host publication	Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop
Pages	1-9
DOIs	https://doi.org/10.18653/v1/2023.ijcnlp-srw.1
Publication status	Published - 2023
Event	Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop - Nusa Dua, Bali Duration: 1 Nov 2023 → 1 Nov 2023

Conference

Conference	Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop
Period	1/11/23 → 1/11/23

Access to Document

10.18653/v1/2023.ijcnlp-srw.1

Cite this

Ghiffari, F. A. A., Alfina, I., & Azizah, K. (2023). Cross-lingual Transfer Learning for Javanese Dependency Parsing. In Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop (pp. 1-9) https://doi.org/10.18653/v1/2023.ijcnlp-srw.1

@inproceedings{7bfd8e837a3f473f8abc42989eb84b35,

title = "Cross-lingual Transfer Learning for Javanese Dependency Parsing",

abstract = "While structure learning achieves remarkable performance in high-resource languages, the situation differs for under-represented languages due to the scarcity of annotated data.This study focuses on assessing the efficacyof transfer learning in enhancing dependency parsing for Javanese—a language spoken by 80million individuals but characterized by limited representation in natural language processing. We utilized the Universal Dependencies dataset consisting of dependency treebanks from more than 100 languages, including Javanese. We propose two learning strategies to train the model: transfer learning (TL) and hierarchical transfer learning (HTL). While TL onlyuses a source language to pre-train the model, the HTL method uses a source language and an intermediate language in the learning process. The results show that our best model usesthe HTL method, which improves performance with an increase of 10 % for both UAS and LAS evaluations compared to the baseline model",

author = "Ghiffari, {Fadli Aulawi Al} and Ika Alfina and Kurniawati Azizah",

year = "2023",

doi = "10.18653/v1/2023.ijcnlp-srw.1",

language = "English",

pages = "1--9",

booktitle = "Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop",

note = "Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop ; Conference date: 01-11-2023 Through 01-11-2023",

}

Ghiffari, FAA, Alfina, I & Azizah, K 2023, Cross-lingual Transfer Learning for Javanese Dependency Parsing. in Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop. pp. 1-9, Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop, 1/11/23. https://doi.org/10.18653/v1/2023.ijcnlp-srw.1

Cross-lingual Transfer Learning for Javanese Dependency Parsing. / Ghiffari, Fadli Aulawi Al; Alfina, Ika ; Azizah, Kurniawati.
Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop. 2023. p. 1-9.

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

TY - GEN

T1 - Cross-lingual Transfer Learning for Javanese Dependency Parsing

AU - Ghiffari, Fadli Aulawi Al

AU - Alfina, Ika

AU - Azizah, Kurniawati

PY - 2023

Y1 - 2023

N2 - While structure learning achieves remarkable performance in high-resource languages, the situation differs for under-represented languages due to the scarcity of annotated data.This study focuses on assessing the efficacyof transfer learning in enhancing dependency parsing for Javanese—a language spoken by 80million individuals but characterized by limited representation in natural language processing. We utilized the Universal Dependencies dataset consisting of dependency treebanks from more than 100 languages, including Javanese. We propose two learning strategies to train the model: transfer learning (TL) and hierarchical transfer learning (HTL). While TL onlyuses a source language to pre-train the model, the HTL method uses a source language and an intermediate language in the learning process. The results show that our best model usesthe HTL method, which improves performance with an increase of 10 % for both UAS and LAS evaluations compared to the baseline model

AB - While structure learning achieves remarkable performance in high-resource languages, the situation differs for under-represented languages due to the scarcity of annotated data.This study focuses on assessing the efficacyof transfer learning in enhancing dependency parsing for Javanese—a language spoken by 80million individuals but characterized by limited representation in natural language processing. We utilized the Universal Dependencies dataset consisting of dependency treebanks from more than 100 languages, including Javanese. We propose two learning strategies to train the model: transfer learning (TL) and hierarchical transfer learning (HTL). While TL onlyuses a source language to pre-train the model, the HTL method uses a source language and an intermediate language in the learning process. The results show that our best model usesthe HTL method, which improves performance with an increase of 10 % for both UAS and LAS evaluations compared to the baseline model

UR - https://aclanthology.org/2023.ijcnlp-srw.1

U2 - 10.18653/v1/2023.ijcnlp-srw.1

DO - 10.18653/v1/2023.ijcnlp-srw.1

M3 - Conference contribution

SP - 1

EP - 9

BT - Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop

T2 - Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop

Y2 - 1 November 2023 through 1 November 2023

ER -

Cross-lingual Transfer Learning for Javanese Dependency Parsing

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this