Design ethical AI systems for nautical crime database

The Food and Agriculture Organization of the United Nations reports that illegal, unreported, and unregulated (IUU) fishing accounts for as much as 26 million tonnes of fish each year, with a value ranging from $10 to $23 billion USD. Despite its significance, there is a notable lack of comprehensive international databases on nautical crimes, including illegal fishing. This gap presents a significant challenge for effectively monitoring and addressing these activities. Advanced machine learning models and large language models (LLMs) are powerful tools for creating a database. However, their use in this context particularly requires extra attention due to its close relation to legal matters and human rights. Our research aims to construct efficient systems to create a nautical crime database while upholding AI ethics and ensuring transparency throughout the process. Our projects are divided into two themes: (1) exploring efficient models to filter relevant articles from web search results and (2) building a system to accurately extract information from the relevant articles.

Faculty Supervisor:

Kristine Bauer

Student:

Partner:

Nautical Crime Investigation Services Corp.

Discipline:

Mathematics

Sector:

Professional, scientific and technical services

University:

University of Calgary

Program: