TWO senior Computer Science students from Silliman University brought Philippine regional language research to the international stage after presenting a pioneering study on the Hiligaynon language at a global conference in Vietnam.

James Ald Y. Teves and Ray Daniel S. Cal of the College of Computer Studies presented their paper, “HiligayNER: A Baseline Named Entity Recognition Model for Hiligaynon,” at the 39th Pacific Asia Conference on Language, Information and Computation held in Hanoi, Vietnam.

The study introduced the first publicly available Named Entity Recognition corpus and baseline models for Hiligaynon, addressing the limited representation of Philippine regional languages in natural language processing research.

The dataset includes more than 8,000 annotated sentences gathered from online news articles, social media posts, and translated texts, providing a valuable resource for low-resource language research.

Using multilingual transformer models mBERT and XLM-RoBERTa, the researchers achieved over 80 percent precision, recall, and F1-score. Cross-language testing also showed potential applications for other Philippine and regional languages.

“The presentation highlights Silliman University’s growing contribution to research on underrepresented Philippine languages and its strong support for student-led scholarship with international impact,” the university said.