A transformer-based prior legal case retrieval method

Simple item page
buir.contributor.authorÖztürk, Ceyhun Emre
buir.contributor.authorÖzçelik, Şemsi Barış
buir.contributor.authorKoç, Aykut
buir.contributor.orcidÖztürk, Ceyhun Emre|0000-0001-9744-6778
buir.contributor.orcidÖzçelik, Şemsi Barış|0000-0002-3666-8366
buir.contributor.orcidKoç, Aykut|0000-0002-6348-2663
dc.citation.epage4en_US
dc.citation.spage1
dc.contributor.authorÖztürk, Ceyhun Emre
dc.contributor.authorÖzçelik, Şemsi Barış
dc.contributor.authorKoç, Aykut
dc.coverage.spatialİstanbul, Türkiye
dc.date.accessioned2024-03-22T12:55:52Z
dc.date.available2024-03-22T12:55:52Z
dc.date.issued2023-08-28
dc.departmentDepartment of Electrical and Electronics Engineering
dc.departmentDepartment of Law
dc.departmentNational Magnetic Resonance Research Center (UMRAM)
dc.descriptionDate of Conference: 05-08 July 2023
dc.descriptionConference Name: 31st IEEE Conference on Signal Processing and Communications Applications, SIU 2023
dc.description.abstractIn this work, BERTurk-Legal, a transformer-based language model, is introduced to retrieve prior legal cases. BERTurk-Legal is pre-trained on a dataset from the Turkish legal domain. This dataset does not contain any labels related to the prior court case retrieval task. Masked language modeling is used to train BERTurk-Legal in a self-supervised manner. With zero-shot classification, BERTurk-Legal provides state-of-the-art results on the dataset consisting of legal cases of the Court of Cassation of Turkey. The results of the experiments show the necessity of developing language models specific to the Turkish law domain.
dc.description.abstractBu çalışmada BERTurk-Legal isimli dönüştürücü tabanlı model emsal karar bulma görevinde kullanılmak üzere önerilmektedir. BERTurk-Legal’in ön eğitimi Türkçe hukuk alanında bir veri kümesi ile yapılmıştır. Bu veri kümesi emsal kararlar ile ilgili herhangi bir etiket bulundurmamaktadır. BERTurk-Legal maskeli dil modelleme kullanılarak kendiliğinden denetimli bir şekilde eğitilmiştir. BERTurk-Legal sınıflandırma görevi üzerinde eğitilmeksizin Yargıtay davalarından oluşan bir veri kümesinde literatürdeki en iyi sonuçları vermiştir. Deney sonuçları Türkçe hukuk alanına özel dil modelleri geliştirme gerekliliğini göstermektedir.
dc.description.provenanceMade available in DSpace on 2024-03-22T12:55:52Z (GMT). No. of bitstreams: 1 A_transformer-based_prior_legal_case_retrieval_method.pdf: 561295 bytes, checksum: 641473ca3ac0af074748d3a3d29fcb24 (MD5) Previous issue date: 2023-08en
dc.identifier.doi10.1109/SIU59756.2023.10223938
dc.identifier.eisbn9798350343557
dc.identifier.isbn9798350343564
dc.identifier.issn2165-0608
dc.identifier.urihttps://hdl.handle.net/11693/115089
dc.language.isoTurkish
dc.publisherIEEE - Institute of Electrical and Electronics Engineers
dc.relation.isversionofhttps://dx.doi.org/10.1109/SIU59756.2023.10223938
dc.source.title2023 31st Signal Processing and Communications Applications Conference (SIU 2023)
dc.subjectNatural language processing
dc.subjectLegal tech
dc.subjectDeep learning
dc.subjectPrior legal case retrieval
dc.subjectLegal NLP
dc.subjectTurkish NLP
dc.subjectDoğal dil işleme
dc.subjectHukuk teknolojileri
dc.subjectDerin öğrenme
dc.subjectEmsal karar bulma
dc.subjectHukukta NLP
dc.subjectTürkçe NLP
dc.titleA transformer-based prior legal case retrieval method
dc.title.alternativeDönüştürücü tabanlı emsal karar bulma yöntemi
dc.typeConference Paper

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
A_transformer-based_prior_legal_case_retrieval_method.pdf
Size:
548.14 KB
Format:
Adobe Portable Document Format
Download

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.01 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Scholarly Publications - Electrical and Electronics Engineering
Scholarly Publications - Law
Scholarly Publications - UMRAM