A Question answering system to answer procedural and causal questions in medicine

• Main Authors: Daella, Paula Angelica, Dimayacyac, Kristina, Liao, Mara Laise, Velarde, Andi
• Format: text
• Language: English
• Published: Animo Repository 2011
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/12095
• Institution: De La Salle University

Existing QA systems on medicine do not answer procedural and effect questions. This research is focused on analyzing, interpreting and answering procedural and causal questions that is commonly raised to medical practitioners in the emergency rooms and in the barrios. The system handles only basic WH-questions such as what, when, why and how. The system also accepts command-type questions using keywords such as List, Give, State and Define. Evaluation showed that the system is able to answer causal questions- with an accuracy of 38.46% wherein the top ranked is correct, 15.38% wherein all answers are correct, 61.54% wherein partial answers are correct. For procedural questions, the system produced and accuracy of 72.73% wherein the top ranked is correct, 72.73% wherein all answers are correct, 90.91% wherein partial answers are correct. Even with need for improvements, medical practitioners recognize the potential of this system as an aid to education and review of students and peers.