Predicting outcomes of the court of cassation of Turkey with recurrent neural networks

buir.contributor.authorÖztürk, Ceyhun E.
buir.contributor.authorÖzçelik, Ş. Barış
buir.contributor.authorKoç, Aykut
buir.contributor.orcidÖztürk, Ceyhun E.|0000-0001-9744-6778
buir.contributor.orcidKoç, Aykut|0000-0002-6348-2663
dc.citation.epage[4]en_US
dc.citation.spage[1]en_US
dc.contributor.authorÖztürk, Ceyhun E.
dc.contributor.authorÖzçelik, Ş. Barış
dc.contributor.authorKoç, Aykut
dc.coverage.spatialSafranbolu, Turkeyen_US
dc.date.accessioned2023-02-15T09:22:27Z
dc.date.available2023-02-15T09:22:27Z
dc.date.issued2022-08-29
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.departmentDepartment of Lawen_US
dc.departmentNational Magnetic Resonance Research Center (UMRAM)en_US
dc.descriptionConference Name: 2022 30th Signal Processing and Communications Applications Conference (SIU)en_US
dc.descriptionDate of Conference: 15-18 May 2022en_US
dc.description.abstractNatural Language Processing (NLP) based approaches have recently become very popular for studies in legal domain. In this work, the outcomes of the cases of the Court of Cassation of Turkey were predicted with the use of Deep Learning models. These models are GRU, LSTM and BiLSTM which are variants of the recurrent neural network. Models saw only fact description parts of the case decision texts during training. Firstly, the models were trained with the word embeddings that were created from the texts from daily language. Then, the models were trained with the word embeddings that were created from downloaded legal cases from Turkish courts. The results of the experiments on the models are given in a comparative and detailed manner. It is observed based on this study and the past studies that the outcomes of the Court of Cassation can be predicted with higher accuracy than most of the courts that were investigated in previous legal NLP studies. The model which is best at predicting decisions is GRU. The GRU model achieves 96.8% accuracy in the decision prediction task.en_US
dc.description.abstractDoğal dil işleme (NLP) tabanlı yaklaşımlar hukuk çalışmalarında son dönemde çok popüler hâle gelmiştir. Bu çalışmada Yargıtay davalarının sonuçları derin öğrenme modelleri kullanılarak tahmin edilmiştir. Bu modeller mükerrer sinir ağı türevi olan GRU, LSTM ve BiLSTM’dir. Modeller eğitim sırasında karar metinlerinin sadece olay tanımları olan kısımlarını görmüştür. İlk olarak modeller günlük dilden metinlerle üretilen kelime temsilleriyle egitilmiştir. Daha sonra modeller Türk mahkemelerinden indirilen davalarla üretilen kelime temsilleriyle eğitilmiştir. Modeller üzerinde yapılan deneylerin sonuçları karşılaştırmalı ve detaylı biçimde verilmiştir. Bu çalışma ve önceki çalışmalara bakılarak Yargıtay kararlarının hukukta yürütülen NLP çalışmalarında incelenen çoğu mahkemeden daha yüksek isabetle tahmin edilebildiği görülmektedir. En başarılı karar tahmini yapan model GRU olarak gözlenmiştir. GRU modeli ile karar tahmininde %96.8 doğruluk skoruna ulaşılmıştır.
dc.description.provenanceSubmitted by Betül Özen (ozen@bilkent.edu.tr) on 2023-02-15T09:22:27Z No. of bitstreams: 1 Predicting_Outcomes_of_the_Court_of_Cassation_of_Turkey_with_Recurrent_Neural_Networks.pdf: 838519 bytes, checksum: a8a82f565099f748269f8c629c4e2d8e (MD5)en
dc.description.provenanceMade available in DSpace on 2023-02-15T09:22:27Z (GMT). No. of bitstreams: 1 Predicting_Outcomes_of_the_Court_of_Cassation_of_Turkey_with_Recurrent_Neural_Networks.pdf: 838519 bytes, checksum: a8a82f565099f748269f8c629c4e2d8e (MD5) Previous issue date: 2022-08-29en
dc.identifier.doi10.1109/SIU55565.2022.9864914en_US
dc.identifier.eisbn978-1-6654-5092-8
dc.identifier.issn2165-0608
dc.identifier.urihttp://hdl.handle.net/11693/111316
dc.language.isoTurkishen_US
dc.publisherIEEEen_US
dc.relation.isversionofhttps://www.doi.org/10.1109/SIU55565.2022.9864914en_US
dc.source.titleSignal Processing and Communications Applications Conference (SIU)en_US
dc.subjectNatural language processingen_US
dc.subjectLawen_US
dc.subjectMachine learningen_US
dc.subjectAI in lawen_US
dc.subjectLegal text miningen_US
dc.subjectDeep learningen_US
dc.subjectDoğal dil işlemeen_US
dc.subjectHukuken_US
dc.subjectMakine öğrenmesien_US
dc.subjectHukukta yapay zekaen_US
dc.subjectHukuki metin işlemeen_US
dc.subjectDerin öğrenmeen_US
dc.titlePredicting outcomes of the court of cassation of Turkey with recurrent neural networksen_US
dc.title.alternativeYargıtay kararlarının mükerrer sinir ağları ile tahminien_US
dc.typeConference Paperen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Predicting_Outcomes_of_the_Court_of_Cassation_of_Turkey_with_Recurrent_Neural_Networks.pdf
Size:
818.87 KB
Format:
Adobe Portable Document Format
Description:

License bundle

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