Judicial Decision Data Gathering, Encoding and Sharing

司法决策数据收集、编码和共享

• 批准号: EP/Y035992/1
• 负责人: David Windridge
• 金额: $ 16.92万
• 依托单位: Middlesex University
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2024
• 资助国家: 英国
• 起止时间: 2024 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FY035992%2F1
• 关键词: Judicial; Decision; Data; Gathering; Encoding

The overall goal of the CHIST-ERA JuDDGES project is to harness state-of-the-art Natural Language Processing & Human-In-The-Loop technologies to provide legal researchers with new Open software and tools that enable extensive, flexible and on-going meta-annotation capability (both automated and employing domain experts in-the-loop). This capability is applied to legal records/judgments from criminal courts across jurisdictions with varied legal constitutions (Poland, England & Wales).We hence seek to dissolve barriers of resources, language, data & format inhomogeneity that currently impede research on judicial decision making. In making this new software, tools and data resource open on a public repository, researchers will be empowered to develop and empirically test theories of judicial decision making and address judicial policy and practice-relevant questions. Researchers from public (legal) institutions can also reuse the data for their purposes. The application of Open Science hence also rectifies a substantial gap in the empirical legal research domain that has been slow in adopting Open Science principles. The resulting annotated pan-national dataset & toolset will constitute the largest and most comprehensive open and reusable legal research repository in Europe for research on judicial decision-making.This project will develop an AI-based solution (tool) that can be used by researchers to examine unstructured textual data in court records and/or written legal judgments. In doing so, we will create the largest extant pan-national legal dataset in Europe. The tool would allow researchers to access and progressively enrich large, detailed, and representative legal samples of data (starting from what we have made available) in a resource- and cost-effective as well as time-efficient way. The datasets so produced and adjusted to Open Science principles and openly available from a trusted data repository, will also be accessible to others for re-use both within and across legal jurisdictions, including the unexpected researcher/user (e.g., from legal institutions). Indeed, the open software & Human-In-The-Loop (HITL) tools that the project will provide will enable non-AI-specialist ECR researchers to interrogate judicial decision data and identify novel lines of research interrogation. The tool and resultant data will thus expand methodological horizons. In the long-term, our project will contribute to a scientific, evidence-based approach to judicial policy and practice in the courts. The project embodies and exemplifies the principles enshrined within Horizon Europe initiatives on open research data, applying Open Science principles and the utilisation of Open Infrastructures.

CHIST-ERA JuDDGES 项目的总体目标是利用最先进的自然语言处理和人机交互技术，为法律研究人员提供新的开放软件和工具，以实现广泛、灵活和在线的元注释功能（自动化和雇用领域专家在环）。此功能适用于具有不同法律宪法的司法管辖区（波兰、英格兰和威尔士）的刑事法院的法律记录/判决。因此，我们寻求消除目前阻碍司法决策研究的资源、语言、数据和格式不均匀性障碍。通过在公共存储库上开放这一新软件、工具和数据资源，研究人员将有权开发和实证检验司法决策理论，并解决司法政策和实践相关问题。公共（法律）机构的研究人员也可以将这些数据用于其目的。因此，开放科学的应用也弥补了实证法律研究领域在采用开放科学原则方面进展缓慢的巨大差距。由此产生的带注释的泛国家数据集和工具集将构成欧洲最大、最全面的开放和可重复使用的法律研究存储库，用于司法决策研究。该项目将开发一个可供研究人员使用的基于人工智能的解决方案（工具）检查法庭记录和/或书面法律判决中的非结构化文本数据。在此过程中，我们将创建欧洲现存最大的泛国家法律数据集。该工具将允许研究人员以资源有效、成本有效以及时间高效的方式访问并逐步丰富大量、详细且具有代表性的法律数据样本（从我们提供的数据开始）。如此生成并根据开放科学原则进行调整并可从可信数据存储库公开获取的数据集也可供其他人访问，以便在法律管辖范围内和跨法律管辖范围内重复使用，包括意外的研究人员/用户（例如，来自法律机构）。事实上，该项目将提供的开放软件和人机交互（HITL）工具将使非人工智能专家 ECR 研究人员能够询问司法判决数据并确定新的研究询问路线。因此，该工具和由此产生的数据将扩大方法论的视野。从长远来看，我们的项目将为法院的司法政策和实践提供科学、循证的方法。该项目体现并例证了 Horizo​​n Europe 倡议中关于开放研究数据、应用开放科学原则和利用开放基础设施的原则。