Costa RicaEducation2020 – ongoingUniversity of Costa Rica
Machine-learning analysis of constitutional judgments
University of Costa Rica researchers used machine learning to examine a collection of more than 500,000 judgments
Problem
Researchers could not read the full volume of constitutional judgments and relied on samples. Inconsistent document formats and subject labels also hindered searches.
Approach
The university's Institute of Legal Research worked with the State of the Nation Program on a project that began by standardizing judgment records. The team implemented thematic classification and automated content analysis using machine learning. Researchers used the tool to find legal arguments and narrow collections to relevant judgments.
Results
Takeaway
Standardizing documents and topic labels preceded automated analysis. The tool remained a research resource without a public search interface.
Categories
- Size
- Large enterprise
- Industry
- Education
- Business function
- IT and R&DFinance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
The university described its judgment-analysis tool and the preparation of the underlying records. Researchers reported examining large collections in minutes; the collection covered more than 500,000 judgments.
Full details are in the original sources.
