CHRExpert: An AI-Driven Court of Human Rights Expert Assistant for Legal Practitioners Utilizing Transformer Models
The growing complexity of human rights litigation, combined with the increasing volume of legal documents, presents significant challenges for legal professionals in efficiently managing case preparation and decision-making. This paper introduces CHRExpert, an Artificial Intelligence (AI)-driven leg...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10909511/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The growing complexity of human rights litigation, combined with the increasing volume of legal documents, presents significant challenges for legal professionals in efficiently managing case preparation and decision-making. This paper introduces CHRExpert, an Artificial Intelligence (AI)-driven legal assistant designed to support law practitioners in analyzing judicial decisions and predicting case outcomes. Utilizing a Pretrained Generative Transformer (GPT) with 6 billion parameters, fine-tuned on the European Court of Human Rights (ECHR) dataset, CHRExpert automates legal document analysis by identifying key legal provisions, performing statutory interpretation, and applying analogical reasoning to suggest legal strategies. The system is designed to support practitioners in aligning their legal arguments with judicial reasoning, particularly in post-litigation analysis. Evaluations based on final judgments demonstrated an accuracy of 83% in predicting outcomes for cases involving Articles 3, 6, and 8 of the European Convention on Human Rights, with an average AUC of 0.93. Additionally, CHRExpert reduced the time required for case preparation by 40%, streamlining legal workflows and improving efficiency. This paper demonstrates the potential practical application of AI to enhance legal processes by offering accurate predictions and experiential insights, making it a valuable tool for human rights litigation. |
|---|---|
| ISSN: | 2169-3536 |