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

The Application of Artificial Intelligence and Legal Analytics: Focused on Decisions Regarding Child Custody

10.6199/NTULJ.201912_48(4).0005

黃詩淳(Sieh-Chuen Huang)；邵軒磊(Hsuan-Lei Shao)

人工智慧 ； 機器學習 ； 法律資料分析 ； 梯度提升法 ； 親權 ； 子女最佳利益 ； 法之可預測性 ； artificial intelligence ； machine learning ； legal analytics ； gradient boosting ； child custody ； best interest of the child ； predictability of law

臺大法學論叢

48卷4期（2019 / 12 / 01）

2023 - 2073

繁體中文

本文試圖應用人工智慧技術至法學問題。首先簡單介紹人工智慧的基本內涵，以及其分支技術機器學習的內容與功能為何，之後說明將其應用於法學研究，具體從事法律資料分析（legal analytics）之實益。其次，本文蒐集2012年1月1日至2014年12月31日共三年期間，當父母均為本國人、也都有意願爭取親權時，地方法院第一審共448件結果為「單獨親權」之裁判，包含了690位子女，使用機器學習當中的「梯度提升法（gradient boosting）」，分析其中民法第1055條之1的各項因素具有多高的重要性。研究發現，最重要的三項因素與比重分別是：主要照顧者為0.356，子女意願為0.267，親子互動為0.152；亦能明確指出其他各項因素的重要性程度。模型準確率（accuracy）為95.7 %，F1分數為0.927，表現相當良好。本文指出了在親權判決中，法官重視的因素以及重視的程度；人工智慧所建構的模型並能有效預測事件的結果，提高裁判的可預測性與透明性。如此，當事人與律師有更充足的資訊評估是否採取訴訟途徑，可能促進訴訟外紛爭解決（調解、和解）的使用率。惟須留意者，由於資料來源的限制，本文僅就公開裁判範圍預測；此外，法院如何判定共同監護、分別監護或第三人監護，亦待將來之研究。

The remarkable advances in artificial intelligence influences human lives in almost every aspect including business and academic research. For example, it became more and more common to use machine learning techniques to analyze, categorize texts and predict outcomes, which can assist human in making more accurate decisions. This research attempts to explore the possibility to apply artificial intelligence approaches to legal studies. Firstly, this study introduces recent developments on artificial intelligence and the basic concepts regarding machine learning. Secondly, it explains how machine learning algorithms can be used to better predict legal outcomes. To demonstrate the strength of predictions, this article applies gradient boosting to analyze decisions related to child custody in Taiwan. We collected 448 cases from 2012 through 2014, involving 690 children whose parents were both Taiwanese and willing to acquire the custody, and in which the Taiwanese district court granted one parent sole custody. It is found that among factors enumerated in Article 1055-1 of Taiwan Civil Code, the three most important ones that judges consider are primary caregiver (gain=0.356), wishes of the child (gain=0.267), and parent-child interaction (gain=0.152). In terms of outcome predictions, the accuracy of the model is 95.7 % and F1 score is 0.927. The model built by gradient boosting could also demonstrate its application on individual cases - that is to say, it is able to reveal factors and how much they weighed on affecting the machine's prediction for a given case. By visualizing through waterfall charts, we may have a better understanding of criteria inside the machine's ＂mind＂. This clearly illustrates in custody disputes what factors and to what extent judges consider important in Taiwan. In addition, this effective predictive model can help improve the predictability and certainly of law. Based on this, divorce lawyers can preliminarily assess their clients' chances at winning divorce lawsuits and propose the most optimal dispute resolution strategy. The informational asymmetries leading to wasteful expenditure on litigation may be reduced. In the long run, legal analytics can improve the acquirability and affordability of information about legal rights and responsibilities, which will enhance public trust and confidence in judicial system.

1. 李立如(2010)。論離婚後父母對未成年子女權利義務之行使負擔：美國法上子女最佳利益原則的發展與努力方向。歐美研究，40(3)，779-828。
   連結：
2. 林琬真,郭宗廷,張桐嘉,顏厥安,陳昭如,林守德(2012)。利用機器學習於中文法律文件之標記、案件分類及量刑預測。中文計算語言學期刊，17(4)，49-67。
   連結：
3. 施慧玲,紀冠伶(2018)。離婚訴訟「先搶先贏」的實務經驗敘事分析：兼論幼年子女最佳利益的司法裁量基準。法令月刊，69(8)，75-102。
   連結：
4. 陳譽文(2017)。人工智慧規範性議題綜觀。科技法律透析，29(4)，43-51。
   連結：
5. 陳顯武,陳世昌(2004)。法資訊學上法本體論研究的興起與發展之分析：一個基本哲學概念意義之轉換。臺大法學論叢，33(5)，1-49。
   連結：
6. 黃詩淳,邵軒磊(2018)。酌定子女親權之重要因素：以決策樹方法分析相關裁判。臺大法學論叢，47(1)，299-344。
   連結：
7. Alarie, B.,Niblett, A.,Yoon, A. H.(2016).Using Machine Learning to Predict Outcomes in Tax Law.Canadian Business Law Journal,58(3),231-254.
8. Ashley, K.,Brüninghaus, S.(2009).Automatically Classifying Case Texts and Predicting Outcomes.Artificial Intelligence and Law,17(2),125-165.
9. Byrd, O.(2017).Moneyball Legal Analytics Now Online for Commercial Litigators.Commercial Law World,31(2),12-16.
10. Chen, C.-J.(2016).The Chorus of Formal Equality: Feminist Custody Law Reform and Fathers’ Rights Advocacy in Taiwan.Canadian Journal of Women and the Law,28(1),116-151.
11. Domingos, P.(2012).A Few Useful Things to Know about Machine Learning.Communications of the ACM,55(10),78-87.
12. Ducklin, P. (2013, October 17). Dirty Dozen Spam Sending Nations: Find Where You Finished in Our Q3 Spampionship Chart [Online forum comment]. Retrieved from https://nakedsecurity.sophos.com/2013/10/17/dirty-dozen-spam-sending-nations-find-where-you-finished-in-our-q3-spampionship-chart/
13. Flach, P. A.(2012).Machine Learning: The Art and Science of Algorithms that Make Sense of Data.New York, NY:Cambridge University Press.
14. Friedman, J. H.(2001).Greedy Function Approximation: A Gradient Boosting Machine.Annals of Statistics,29(5),1189-1232.
15. Hawkins, D. M.,Basak, S. C.,Mills, D.(2003).Assessing Model Fit by Cross-Validation.Journal of Chemical Information and Computer Sciences,43(2),579-586.
16. Karamouzis, S. T.,Harper, D. W.(2008).An Artificial Intelligence System Suggests Arbitrariness of Death Penalty.International Journal of Law and Information Technology,16(1),1-7.
17. Katz, D. M.(2013).Quantitative Legal Prediction - or - How I Learned to Stop Worrying and Start Preparing for the Data-Driven Future of the Legal Services Industry.Emory Law Journal,62,909-966.
18. Kim, J.-H.(2009).Estimating Classification Error Rate: Repeated Cross-Validation, Repeated Hold-Out and Bootstrap.Computational Statistics and Data Analysis,53(11),3735-3745.
19. Kuhn, M.,Johnson, K.(2013).Applied Predictive Modeling.New York, NY:Springer.
20. Le, Q. V.,Ranzator, M’A.,Monga, R.,Devin, M.,Chen, K.,Corrado, G. S.,Dean, J.,Ng ,A. Y.(2012).Building High-Level Features Using Large Scale Unsupervised Learning.Proceedings of the 29th International Conference on Machine Learning,Madison, WI:
21. Maisel, L.,Cokins, G.(2014).Predictive Business Analytics: Forward Looking Capabilities to Improve Business Performance.Hoboken, NJ:Wiley.
22. Manning, C. D.,Raghavan, P.,Schütze, H.(2008).Introduction to Information Retrieval.New York, NY:Cambridge University Press.
23. Marsland, S.(2014).Machine Learning: An Algorithmic Perspective.Boca Raton, FL:Chapman and Hall/CRC.
24. McGinnis, J. O.,Pearce, R. G.(2014).The Great Disruption: How Machine Intelligence Will Transform the Role of Lawyers in the Delivery of Legal Services.Fordham Law Review,82(6),3041-3066.
25. McShane, B. B.,Watson, O. P.,Baker, T.,Griffith, S. J.(2012).Predicting Securities Fraud Settlements and Amounts: A Hierarchical Bayesian Model of Federal Securities Class Action Lawsuits.Journal of Empirical Legal Studies,9(3),482-510.
26. Molinaro, A. M.,Simon, R.,Pfeiffer, R. M.(2005).Prediction Error Estimation: A Comparison of Resampling Methods.Bioinformatics,21,3301-3307.
27. Raghupathi, V.,Zhou, Y.,Raghupathi, W.(2018).Legal Decision Support: Exploring Big Data Analytics Approach to Modeling Pharma Patent Validity Cases.IEEE Access,6,41518-41528.
28. Ramseyer, J. M.(2015).Second-Best Justice: The Virtues of Japanese Private Law.Chicago, IL:The University of Chicago Press.
29. Russell, S. J.(Ed.),Norvig, P.(Ed.)(2009).Artificial Intelligence: A Modern Approach.Upper Saddle River, NJ:Pearson.
30. Sammut, C.(Ed.),Webb, G. I.(Ed.)(2017).Encyclopedia of Machine Learning and Data Mining.Boston, MA:Springer US.
31. Siegel, E.(2013).Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die.Hoboken, NJ:Wiley.
32. Sorkin, D. E.(2001).Technical and Legal Approaches to Unsolicited Electronic Mail.University of San Francisco Law Review,35(2),325-384.
33. Witten, I. H.,Frank, E.,Hall, M. A.(2011).Data Mining: Practical Machine Learning Tools and Techniques.Burlington, MA:Morgan Kaufmann Publishers.
34. 中川淳(2008)。婚姻費用分担額と標準的算定方式：最判平成 18.4.26 を中心として。戸籍時報，622，62-65。
35. 王子榮（2018），〈恐龍法官製造公司：當戰場不在法庭〉，聯合新聞網鳴人堂版，載於：https://opinion.udn.com/opinion/story/11756/2930248
36. 林沛君(2015)。由聯合國兒童權利委員會第 14 號一般性意見重新檢視「子女最佳利益」。華岡法粹，58，127-159。
37. 林勤富,劉漢威(2018)。人工智慧法律議題初探。月旦法學雜誌，274，195-215。
38. 林勤綱(1982)。臺北，國立臺灣大學法律學研究所。
39. 張永健(2017)。社會科學式的比較法研究：評 Mark Ramseyer. 2015. Second Best Justice: The Virtues of Japanese Private Law. The University of Chicago Press。中研院法學期刊，20，211-250。
40. 張永健,何漢葳,李宗憲(2017)。或重於泰山、或輕於鴻毛：地方法院車禍致死案件慰撫金之實證研究。政大法學評論，149，139-219。
41. 郭雨嵐,汪家倩,侯春岑(2017)。法律科技與人工智慧時代，科技法律人才的養成與挑戰。萬國法律，214，51-59。
42. 黃詩淳,邵軒磊(2017)。運用機器學習預測法院裁判：法資訊學之實踐。月旦法學雜誌，270，86-96。
43. 劉宏恩(2011)。「子女最佳利益原則」在臺灣法院離婚後子女監護案件中之實踐：法律與社會研究（Law and Society Research）之觀點。軍法專刊，57(1)，84-106。
44. 鄭諺霓(2015)。臺北，國立臺灣大學法律學研究所。
45. 鄧學仁(2011)。離婚後子女親權酌定之問題與對策。月旦法學雜誌，191，34-44。
46. 鄧學仁(2010)。子女最佳利益之適用爭議與發展方向。台灣法學雜誌，155，45-61。
47. 舘幸嗣、打矢恵(2007)。扶養義務と養育費算定に関する一考察。中央学院大学法学論叢，20(1?2)，1-22。

1. 洪德欽(2020)。歐盟自動駕駛車之發展策略與法律規範。歐美研究，50(2)，349-431。
2. 黃詩淳(2023)。AI可解釋性的法學意義及其實踐。臺大法學論叢，52(S)，931-972。
3. 黃詩淳,邵軒磊(2020)。新住民相關親權酌定裁判書的文字探勘：對「平等」問題的法實證研究嘗試。臺大法學論叢，49(S)，1267-1308。
4. 黃詩淳,邵軒磊(2020)。以人工智慧讀取親權酌定裁判文本：自然語言與文字探勘之實踐。臺大法學論叢，49(1)，195-224。
5. (2020)。初探車禍判決中法院認定之過失比例之因素。月旦法學雜誌，305，206-221。
6. (2020)。論人工智慧技術應用、法律問題定位及監管立法趨勢──以美國實務發展為核心。成大法學，39，173-235。
7. (2022)。智慧法院之發展與界限（上）──演算法、科技治理與司法韌性。月旦法學雜誌，323，72-98。
8. (2023)。我國法實證研究社群的發展現況—知識結構、引用網絡與質性分析。中研院法學期刊，33，1-80。