EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Artificial Intelligence–Driven Legal Decision Support Systems: Emerging Frameworks for Technology Law, Cyber Law, and Digital Governance

Dr Shatakshi Anand, Dr. Pushkar Shankar Shukla, Jhalak Nandwani, Dr.Taru Mishra, Dr Pawan Kumar, Aman Kumar

Computer and Decision Making: An International Journal · 06.11.2026 · 3 (8S) · 13-21

DOI: 10.1922/comdem.v3i8s.19111 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 06.11.2026. Verfasst von Dr Shatakshi Anand und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The complexity of regulatory texts complicates the assessment of whether organisational texts reveal what is required by data-protection law. The present research investigates the potential of supervised text classification for helping human review of GDPR related disclosure in privacy policies. With the annotated dataset of Al Rahat, Long and Tian, a leakage controlled sample of 7,918 unique text segments from 10,511 original records was created. The sample was split into three sets: training, validation, and test, based on domain of the website. Multinomial Naive Bayes and two Logistic Regression models (word-level and combined word- and character-level TF-IDF) have been tested in one-versus-rest multi-label classification of 18 GDPR disclosure categories. Validation data was used for tuning thresholds for each category. On the held-out domains, the selected combined-feature Logistic Regression obtained macro-F1 = 0.8347, micro-F1 = 0.8439, and exact-match accuracy = 75.28%. A 95% confidence interval for macro-F1 using the domain-cluster bootstrap was (0.8022, 0.8621). It was only marginally better and not statistically significant than word-only Logistic Regression. There was some variation in performance by categories and multi-label segments performed lower. A disclosure-category classification task has been validated, not a legal-compliance task. Such classifiers are introduced in a conceptual and unevaluated framework in the context of human-supervised legal decision support.

Zitieren

APA

Anand, D. S., Shukla, D. P. S., Nandwani, J., Mishra, D., Kumar, D. P., & Kumar, A. (2026). Artificial Intelligence–Driven Legal Decision Support Systems: Emerging Frameworks for Technology Law, Cyber Law, and Digital Governance. Computer and Decision Making: An International Journal, 3(8S), 13-21. https://doi.org/10.1922/comdem.v3i8s.19111

Vancouver

Anand DS, Shukla DPS, Nandwani J, Mishra D, Kumar DP, Kumar A. Artificial Intelligence–Driven Legal Decision Support Systems: Emerging Frameworks for Technology Law, Cyber Law, and Digital Governance. Computer and Decision Making: An International Journal. 2026;3(8S):13-21. doi:10.1922/comdem.v3i8s.19111

BibTeX
@article{Anand2026,
  author = {Anand, Dr Shatakshi and Shukla, Dr. Pushkar Shankar and Nandwani, Jhalak and Mishra, Dr.Taru and Kumar, Dr Pawan and Kumar, Aman},
  title = {Artificial Intelligence–Driven Legal Decision Support Systems: Emerging Frameworks for Technology Law, Cyber Law, and Digital Governance},
  journal = {Computer and Decision Making: An International Journal},
  year = {2026},
  volume = {3},
  number = {8S},
  pages = {13--21},
  doi = {10.1922/comdem.v3i8s.19111}
}
Kostenlos starten →