• Sherman Ho and Ellie Tse

Hong Kong students use machine learning to help unrepresented litigants read legislation - Decoding

The number of unrepresented litigants in Hong Kong has stayed high in recent years. According to a recent Legislative Council report, nearly half of the hearings in Hong Kong courts involve unrepresented, "Litigants in Persons". In other words, a significant amount of laymen unfamiliar with the law and relevant procedures are pitted against experienced lawyers. It does not help that despite most cases are tried in Chinese, the Chinese version of the legislation is, as Yeung observed in a recent study, “quite unreadable”. Given the above, unrepresented litigants are placed at a tremendous disadvantage in the courts of Hong Kong.

In light of this, a group of law students from the Chinese University of Hong Kong and the University of Hong Kong launched an initiative they have called “Decoding Law” to enhance access to justice. The project aims to utilize machine learning to aid unrepresented litigants under the current system.

Breaking down legislations into edible pieces

A survey done recently shows that over 70% of the unrepresented litigants had difficulties in understanding the applicable legislation. In response to this, Decoding Law has created a web browser (Chrome) plug-in to break down legislation into easily digestible pieces. The plug-in constitutes three features, namely (1) chatbot, (2) definition bubbles and (3) sentence structure analysis.

When a user types a question into the chatbot, it will identify possible relevant ordinance sections and trusted online resources for users. For example, when users type in “Is playing mahjong illegal?”, the chatbot will generate an answer that shows “Cap. 148 Gambling Ordinance section 3”, the relevant section(s) and “Unlawful Gambling - Senior CLIC”, the trusted online resource (see Illust. One). The answer generated is more specific and reliable than a simple search in Google, which may include multiple contradictory answers and resources from biased or unreliable sources.

The Decoding Law Chatbot

After a user has accessed the relevant sections via the chatbot or otherwise, the legally defined terms will be highlighted in yellow. When a user hovers her cursor over the highlighted words, a definition bubble will pop up showing the legal definition of that term, as illustrated below. With this feature, users do not have to scroll back and forth between sections, and thus are able better to grasp the meaning of the legislation in a more efficient manner.

Help defining legal terms

If users have trouble reading some sections (even legal professionals occasionally do), they can click on the "Analysis" button located around those sections. The plugin will break down the sections into three categories, namely “conditions”, “results” and “exceptions”. This should ease the difficulty and make it more manageable for users to read convoluted sections.

Analysis of Content

More than just an idea

In April, Decoding Law won the public sector award at the inaugural Global Legal Hackathon 2018 in New York. The competitors included practicing lawyers and professionals from different sectors, coming from 40 cities across 22 countries worldwide. Decoding Law was subsequently featured in the South China Morning Post, RTHK, and legal publications such as LexisNexis’ the Lawyers Daily and ABA Journal. While the students may be among the first to experiment with innovative ideas, there is no doubt that the legal industry is gradually picking up its pace to embrace technology for a better future.

Sherman Ho and Ellie Tse, Hong Kong

100 views0 comments

Recent Posts

See All

Judging Autonomous Vehicles

Jeffrey J. Rachlinski – Cornell Law School; Andrew J. Wistrich – California Central District Court -- Would you rather be run over by a self-driving car or a car driven by a human being? Assuming a s

What is Legal Innovation?

Haim Sandberg – School of Law, The College of Management Academic Studies -- Is law itself an arena of innovation – of legal innovation? Can we learn something about the nature of legal innovation fr

Measuring Law Over Time

Corinna Coupette, Max Planck Institute for Informatics; Janis Beckedorf, Ruprecht-Karls-Universität Heidelberg; Dirk Hartung, Center for Legal Technology and Data Science; Michael Bommarito, CodeX – T