메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(Seoul National University of Science and Technology) (Seoul National University of Science and Technology)
저널정보
한국HCI학회 한국HCI학회 학술대회 PROCEEDINGS OF HCI KOREA 2023 학술대회 발표 논문집
오류 신고하기

피인용 0

검색

    초록·키워드

    Narrative artificial intelligence (AI), which adds narrative persuasion to simple causal interpretation, is an explainable AI technique that helps patients understand their health data (e.g., MRI results). Narrative AI is expected to improve causability for health data and thereby induce meaningful behavioral change. However, research on which type of narrative AI (i.e., counterfactual or prefactual) better improves causability for complex health data is still elusive. This study created two different types of narrative AI for MRI reports (i.e., counterfactual and prefactual) and compared their impact on causability with 20 participants. System causability scale results showed that both narrative AI provided a high level of causability. Semi-structured interview results showed that the two narrative AIs improve causability by different mechanisms. While counterfactual made participants focus on self-reflection, prefactual made participants seek self-improvement to prepare for the future. Our findings showed that the two types of narrative AI (i.e., counterfactual and prefactual) improve causability by different mechanisms. By understanding these different underlying mechanisms, we expect to be able to design and deliver more personalized narrative AI to users.

    최근 본 자료 전체보기