{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T16:22:57Z","timestamp":1774369377807,"version":"3.50.1"},"reference-count":0,"publisher":"IOS Press","isbn-type":[{"value":"9781643686387","type":"electronic"}],"license":[{"start":{"date-parts":[[2025,12,2]],"date-time":"2025-12-02T00:00:00Z","timestamp":1764633600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,12,2]]},"abstract":"<jats:p>Algorithmic bias in bail decision systems threatens judicial fairness, yet research on computational analysis of Indian court judgments remains scarce. Building on the recently released IndianBailJudgments-1200 dataset [4], which includes 1,200 bail cases from Indian High Courts and the Supreme Court (1975\u20132025) with rich textual and structured socio-legal attributes, we present the first comprehensive empirical study addressing both bail outcome prediction and fairness evaluation in this context. For bail outcome prediction, we benchmark Logistic Regression and domain-adapted LegalBERT, leveraging combined textual and structured features. For fairness evaluation, we compare Logistic Regression, LegalBERT, and XGBoost across gender, crime type, and region subgroups. Logistic Regression achieves 95.4% baseline accuracy but suffers data leakage (drops to 88.8% after correction); LegalBERT reaches 86.9% with greater robustness. In fairness evaluation, XGBoost attains 95.6% accuracy with lowest gender disparity (1.8%), while LegalBERT achieves 93.9% and Logistic Regression 93.1%. Critically, data leakage correction reveals severe fairness degradation in transformers, with LegalBERT dropping to 82.2%. We will release models, code, and splits, establishing the first reproducible fairness benchmark for Indian bail judgments, and advocate for fairness-aware, human-supervised AI in judicial settings.<\/jats:p>","DOI":"10.3233\/faia251599","type":"book-chapter","created":{"date-parts":[[2025,12,4]],"date-time":"2025-12-04T12:05:08Z","timestamp":1764849908000},"source":"Crossref","is-referenced-by-count":1,"title":["From Data to Equity: Predicting and Auditing Fairness in Indian Bail Decisions"],"prefix":"10.3233","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-6062-2863","authenticated-orcid":false,"given":"Sneha","family":"Deshmukh","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, Datta Meghe College of Engineering, Maharashtra, India"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-0306-8003","authenticated-orcid":false,"given":"Prathmesh","family":"Kamble","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Datta Meghe College of Engineering, Maharashtra, India"}]}],"member":"7437","container-title":["Frontiers in Artificial Intelligence and Applications","Legal Knowledge and Information Systems"],"original-title":[],"link":[{"URL":"https:\/\/ebooks.iospress.nl\/pdf\/doi\/10.3233\/FAIA251599","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,4]],"date-time":"2025-12-04T12:05:09Z","timestamp":1764849909000},"score":1,"resource":{"primary":{"URL":"https:\/\/ebooks.iospress.nl\/doi\/10.3233\/FAIA251599"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,2]]},"ISBN":["9781643686387"],"references-count":0,"URL":"https:\/\/doi.org\/10.3233\/faia251599","relation":{},"ISSN":["0922-6389","1879-8314"],"issn-type":[{"value":"0922-6389","type":"print"},{"value":"1879-8314","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,2]]}}}