政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/144042
English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  全文笔数/总笔数 : 110206/141131 (78%)
造访人次 : 46889111      在线人数 : 652
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜寻范围 查询小技巧:
  • 您可在西文检索词汇前后加上"双引号",以获取较精准的检索结果
  • 若欲以作者姓名搜寻,建议至进阶搜寻限定作者字段,可获得较完整数据
  • 进阶搜寻
    政大機構典藏 > 資訊學院 > 資訊科學系 > 學位論文 >  Item 140.119/144042


    请使用永久网址来引用或连结此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/144042


    题名: 中文司法裁判文書標記輔助環境初探
    A Prototype for Assisting the Labeling of Judicial Documents in Chinese
    作者: 黃翊唐
    Huang, Yi-Tang
    贡献者: 劉昭麟
    黃翊唐
    Huang, Yi-Tang
    关键词: 法學資訊
    系統開發
    自然語言處理
    判決書
    Legal informatics
    System development
    Natural language processing
    Judicial ruling
    日期: 2023
    上传时间: 2023-04-06 18:00:07 (UTC+8)
    摘要:   隨著科技日新月異,許多產業導入各種硬軟體進行自動化與數位化,但是在法學資訊領域上這件事情較難以發展,原因可能是案發原因複雜,難以使用文字完整記載、法官判決之理由也不全然會寫在判決書內,所以法學資訊相較於其他領域較難以發展。我們希望透過將判決書的各種標記,把判決書中的某些類別標記出來,像是:爭點、法官見解等等,將判決書中的線索解構出來,方便進行後續的檢索甚至是機器學習等等應用,本篇論文在研究標記系統的開發方法與相關技術。
      由於訓練品質精良的機器學習與深度學習模型,仰賴極大量的資料對模型進行訓練及測試,這些資料都必需由人工標記,極大量的資料透過人工瀏覽與標記是很容易出錯的,為了降低標記的錯誤率,以及提升資料集的品質,我們希望透過開發一套判決書的標記輔助系統,使用者能夠在本系統上進行判決書的檢索、瀏覽、上傳、標記與下載標記成果,藉由降低標記的難度,以及提升工作流程的順暢度,達到降低錯誤率的結果。
      As technology advances, many industries are adopting various software and hardware for automation and data management, but it is difficult to develop in the field of legal informatics. This may be because the reasons for the case are complicated and difficult to fully record in writing, and the reasons for the judge`s ruling are not entirely written in the ruling. Therefore, legal informatics is difficult to develop compared to other fields. We hope to mark the various marks in the ruling, mark out some categories in the ruling, such as: points of contention, judge`s opinion, etc., and deconstruct the clues in the ruling to facilitate subsequent retrieval or even machine learning applications. This paper studies the development methods and related technologies of the labeling system.
      Due to the training and deep learning models with excellent training quality, it depends on a large amount of data to train and test the model. These materials must be manually marked by manual. The error rate and the quality of the data set, we hope to develop a set of judgment auxiliary systems, users can retrieve, browsing, uploading, marking, and download markers on this system to reduce the marking marks. Difficulty, and improve the smoothness of the workflow to reduce the result of reducing error rates.
    參考文獻: 司法院法學資料檢索系統 (2023)。檢自https://lawsearch.judicial.gov.tw/(January 01, 2023)
    劉一凡、劉昭麟及楊婕。以民事訴訟之爭點分群為基礎的類似案件搜尋系統(Clustering Issues in Civil Judgments for Recommending Similar Cases),第卅四屆自然語言與語音處理研討會論文集 (ROCLING XXXIV),184-192。2022。
    劉威志、林泓任、吳柏憲及劉昭麟。老年扶養費請求案件之准駁及扶養金額
    預測 (Predicting judgments and grants for civil cases of alimony for the elderly), 第卅四屆自然語言與語音處理研討會論文集 (ROCLING XXXIV),121-128。2022。
    林泓任、劉威志、劉昭麟及楊婕。以機器學習與規則方法辨識中文民事裁判
    書結構 (Using machine learning and pattern-based methods for identifying elements in Chinese judgment documents of civil cases),第卅四屆自然語言與語音處理研討會論文集 (ROCLING XXXIV),107-115。2022。
    Elasticsearch. Retrieved from https://www.elastic.co/ (January 01, 2023)
    黃詩淳、邵軒磊,人工智慧與法律資料分析之方法與應用:以單獨親權酌定裁判的預測模型為例。臺大法學論叢,第 48 卷第 4 期,2023-2073。2019。
    司 法 院 : 國 民 法 官 制 度 (2023)。檢自https://social.judicial.gov.tw/CJlandingpage/ (February 03, 2023)
    Label Studio. Retrieved from https://labelstud.io/ (January 01, 2023)
    Pip. Retrieved from https://pypi.org/project/pip/ (January 01, 2023)
    Amazon Web Services. Retrieved from https://aws.amazon.com/ (February 03, 2023)
    Google Cloud Platform. Retrieved from https://console.cloud.google.com/ (February 03, 2023)
    Microsoft Azure. Retrieved from https://azure.microsoft.com/ (February 03, 2023)
    MARKUS. Retrieved from https://dh.chinese-empires.eu/markus/beta/index.html (January 01, 2023)
    MedKnowts. Retrieved from http://clinicalml.org/projects/medknowts/ (February 03, 2023)
    Murray, Luke, et al. Medknowts: unified documentation and information retrieval for electronic health records. The 34th Annual ACM Symposium on User Interface Software and Technology. 2021.
    Chen, Irene Y., Rahul G. Krishnan, and David Sontag. Clustering Interval-Censored Time-Series for Disease Phenotyping. Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 36, No. 6, 6211-6221. 2022.
    Karlsson, Rickard KA, et al. Using time-series privileged information for provably efficient learning of prediction models. arXiv preprint arXiv:2110.14993. 2021.
    Huang, Yi-Tang, Hong-Ren Lin, and Chao-Lin Liu. Toward an Integrated Annotation and Inference Platform for Enhancing Justifications for Algorithmically Generated Legal Recommendations and Decisions. Legal Knowledge and Information Systems. IOS Press, 281-285. 2022.
    描述: 碩士
    國立政治大學
    資訊科學系
    108753132
    資料來源: http://thesis.lib.nccu.edu.tw/record/#G0108753132
    数据类型: thesis
    显示于类别:[資訊科學系] 學位論文

    文件中的档案:

    档案 描述 大小格式浏览次数
    313201.pdf3584KbAdobe PDF20检视/开启


    在政大典藏中所有的数据项都受到原著作权保护.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本網站維護人員(nccur@nccu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回馈