본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색
결과 내 검색 동의어 포함
결과 내 검색 동의어 포함

-

초록보기

This study aimed to create machine learning-based screening tools for middle school students to predict suicide planning or attempts in school settings, and to identify the risk factors. A total of 3,812 records from a national database established through rigorous sampling designs, the 2022 Korean Youth Risk Behavior Web-based Survey (KYRBWS), were used to ensure high generalizability and minimize bias of the study’s outcomes. Five machine-learning models, utilizing various algorithms available in SAS or Python, underwent training with a 10-fold validation process. Performance evaluation included metrics such as accuracy, sensitivity, specificity, and AUC, resulting from using a common unseen test dataset. Feature importance was interpreted through the average absolute values of Shapley additive explanations (SHAP) as well as the standardized regression coefficients. The artificial neural-network model with three hidden layers and dropout layers between each dense layer emerged as the top performer. It achieved sensitivity and AUC exceeding 91%, with accuracy and specificity at 86%. However, all five models, utilizing an optimal set of nine features, demonstrated strong performance. Mental health features such as suicidal thoughts, feeling lonely, or feeling sad, and the subjective perception of dental health emerged as significant risk factors. Adolescents considering suicide were approximately 12.7 times more likely to make plans or attempt suicide than their peers without such ideation (OR = 12.7; 95% CI = [8.0-20.1]). Individuals perceiving dental health as 'Very bad' were 4.4 times more prone to suicidal behavior than those with 'Average' ratings (OR = 4.4; 95% CI = [1.4-13.5]). Those always feeling lonely were 2.8 times more likely to engage in suicidal behavior (OR = 2.8; 95% CI = [1.1-7.1]), while persistent feelings of sadness increased the likelihood by 1.8 times (OR = 1.8; 95% CI = [1.2-2.8]). The study also offers valuable guidance on algorithm selection and suitable options for similar applications.

권호기사

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 목차
중등 교사를 위한 미래문제해결 프로그램 기반 정보, 과학, 수학 교과의 인공지능 융합교육 연수 교육과정 개발 및 효과 분석 = Development and effect analysis of future problem solving program based informatics, mathematics, science convergence education teachers' training course using AI for secondary school teachers 이다겸, 이영준, 백성혜 p. 1-11

생성 AI를 활용한 NCS 직무능력 평가를 위한 자동문항생성 방법론 연구 = A study on the automatic generation methodology of NCS-based job competency assessment items using generative AI : focused on GPT4-o based information security analysis competency unit : GPT4-o 기반 정보보호 분야 침해사고 분석 능력단위 중심 이재식 p. 13-25

생성형 AI를 활용한 협력학습 과정에서 나타나는 상호작용 양상 탐색 = Comparison of the interaction patterns in collaborative learning using generative AI 박하나 p. 27-34

중학교 수학 기반 AI융합교육 프로그램의 개발과 적용 = Development and application of AI convergence education program focus on mathematics in middle school 김진희, 김동심 p. 35-46

딜레마 토론 기반 인공지능 윤리교육 모델 개발 = Development of AI ethics education model based on dilemma discussion 김은경, 이영준 p. 47-56

AI 디지털교과서 학습맵에 기반한 정보 교과 진단평가 문항 개발 = Development of informatics subject diagnostic assessment questions based on AI digital textbook learning map 이정숙, 최현종 p. 57-66

Screening adolescent suicidal behavior and risk factors : a machine-learning approach Seon-Hi Shin, Bossng Kang, Myung-Suk Woo, Youn-Ju Park p. 67-78

메타버스 ZEP을 활용한 청소년 대상 자기공감 프로그램 개발 = Development of a self-empathy program targeted at adolescents using metaverse ZEP 최선영, 김소영, 강정애 p. 79-88

2022 개정 초등 교육과정 성취기준에 반영된 데이터 리터러시 지식의 깊이(Depth of Knowledge) 분석 = An analysis of the depth of knowledge in data literacy reflected in the 2022 revised elementary curriculum achievement standards 문현우, 손정명, 이시훈, 이영준 p. 89-100

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차