When defendants speak: Quantifying the predictive value of defence arguments in construction litigation
Yazarlar (3)
Dr. Öğr. Üyesi Mahmut SARI Kırşehir Ahi Evran Üniversitesi, Türkiye
Savaş Bayram Erciyes Üniversitesi, Türkiye
Emrah Aydemir Sakarya Üniversitesi, Türkiye
| Makale Türü |
Özgün Makale
(ESCI dergilerinde yayınlanan tam makale)
|
||
| Dergi Adı | Journal of Construction Engineering, Management & Innovation | ||
| Dergi ISSN | 2630-5771 | ||
| Dergi Tarandığı Indeksler | ESCI | ||
| Makale Dili | İngilizce | Basım Tarihi | 03-2025 |
| Cilt / Sayı / Sayfa | 8 / 1 / 64–88 | DOI | 10.31462/jcemi.2025.01064088 |
| Makale Linki | https://doi.org/10.31462/jcemi.2025.01064088 | ||
| Özet |
| Artificial Intelligence (AI) has catalysed a paradigm shift in legal analytics, enabling datadriven interrogation of judicial texts across jurisdictions [1, 2]. Yet, despite these advancements, the administration of justice remains mired in inefficiency—case backlogs, spiralling litigation durations, and eroding public trust plague courts globally [3, 4]. In Turkey, where the Civil Chambers of the Court of Cassation saw an 8.6% rise in average case duration (221 days in 2023 to240 days in 2024)[5], the crisis underscores the unsustainability of traditional legal practices. Legal Judgment Prediction (LJP), which leverages Natural Language Processing (NLP) to automate outcome forecasting via case fact analysis [6], offers transformative potential. However, its application to agglutinative languages like Turkish and sector-specific disputes, such as construction, remains critically underexplored. The Turkish judiciary’s hierarchical structure—comprising Courts of First Instance, Regional Courts, and the precedent-setting Court of |
| Anahtar Kelimeler |
| Court of cassation | Natural language processing | Judicial decision prediction | Text classification | Machine learning |
BM Sürdürülebilir Kalkınma Amaçları
| Atıf Sayıları | |
| Google Scholar | 1 |
| Web of Science | 2 |