Bi-directional English-Hiligaynon Statistical Machine Translation.


Click here for the link to the paper

The paper was presented at the 2017 IEEE Region 10 Conference (TENCON 2017)

TENCON 2017 is expected to bring together researchers, educators, students, practitioners, technocrats and policymakers from across academia, government, industry and non-governmental organizations to discuss, share and promote current works and recent accomplishments across all aspects of electrical, electronic and computer engineering, as well as information technology. Distinguished people will be invited to deliver keynote speeches and invited talks on trends and significant advances in the emerging technologies.


  • Dana Genevieve Macabante
  • John Casper Tambanillo
  • Angelica Dela Cruz
  • Nove Ellema
  • Manolito Octaviano Jr.
  • Ramon Rodriguez
  • Rachel Edita Roxas


In this study, we implemented a bi-directional English-Hiligaynon translator using Statistical Machine Translation framework. Hiligaynon is a member of the Malayo-Polynesian language family and is the fourth major language in the Philippines. Text documents from the New Testament were used as training data. These were cleaned and tokenized, and a parallel corpus was developed. To evaluate the performance of the translator, Bilingual Evaluation Understudy (BLEU) Score was used. We were able to achieve 21.74 for Hiligaynon to English and 24.43 for English to Hiligaynon. Philippine languages have free-word order, which would explain why we were able to get a lower score for Hiligaynon to English. For future work, increasing the size of the training data and using deep learning can be explored.

Posts created 35

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts

Begin typing your search term above and press enter to search. Press ESC to cancel.

Back To Top