RISS 학술연구정보서비스

검색

인기 검색어

    다국어 입력

    http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

    변환된 중국어를 복사하여 사용하시면 됩니다.

    예시)
    • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
    • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
    닫기

    머신러닝과 통계적 방법을 이용한 대사성 질환과 식품 소비와의 관계성 연구

    한글로보기

    https://www.riss.kr/link?id=T16060230

    • 저자
    • 발행사항

      청주 : 충북대학교, 2022

    • 학위논문사항

      학위논문(석사) -- 충북대학교 , 빅데이터 협동과정 , 2022. 2

    • 발행연도

      2022

    • 작성언어

      한국어

    • 주제어
    • KDC

      005.74 판사항(5)

    • 발행국(도시)

      충청북도

    • 기타서명

      A Study on Relationship between Metabolic Diseases and Food Consumption using Machine Learning and Statistical Method

    • 형태사항

      v,42p. : 삽화, 표 ; 26cm

    • 일반주기명

      충북대학교 논문은 저작권에 의해 보호됩니다
      지도교수:조완섭
      참고문헌 : p. 40-42

    • UCI식별코드

      I804:43009-000000056806

    • 소장기관
      • 충북대학교 도서관 소장기관정보
    • 0

      상세조회
    • 0

      다운로드
    서지정보 열기
    • 내보내기
    • 내책장담기
    • 공유하기
    • 오류접수

    부가정보

    다국어 초록 (Multilingual Abstract) kakao i 다국어 번역

    Abstract

    Food is not only a source of nutrients such as carbohydrates, proteins, and lipids, but also various physiologically active ingredients contained in food can contribute to preventing diseases through various biological functions regulating actions. This research aims to identify the relationship between food and specific diseases, combining consumer panel data from RDA, which contains information on purchasing agri-food by household, and health checkup and medical treatment data from NHIS. The patient group was divided into 52 patients diagnosed with metabolic disease, and the control group was divided into 276 patients without metabolic disease diagnosis. In order to examine the relationship between food consumption and disease, we compared the monthly food purchases of the patient group before the disease occurred and the food purchases of the control group. It was determined that the interaction existed at the significance level α=0.1 as a result of testing the effect of two-way ANOVA interaction between the variable dividing the patient group and the control group and the purchased food. In addition, as a result of the post-hoc test, there was an interaction effect between the patient group and control group in fruits, livestock products, aquatic products, seasoned vegetables, confectionery, and fruits and vegetables. Moreover, there was a difference in food consumption patterns between the patient group and the control group.

    Key words :
    번역하기

    Abstract Food is not only a source of nutrients such as carbohydrates, proteins, and lipids, but also various physiologically active ingredients contained in food can contribute to preventing diseases through various biological functions regulating...

    Abstract

    Food is not only a source of nutrients such as carbohydrates, proteins, and lipids, but also various physiologically active ingredients contained in food can contribute to preventing diseases through various biological functions regulating actions. This research aims to identify the relationship between food and specific diseases, combining consumer panel data from RDA, which contains information on purchasing agri-food by household, and health checkup and medical treatment data from NHIS. The patient group was divided into 52 patients diagnosed with metabolic disease, and the control group was divided into 276 patients without metabolic disease diagnosis. In order to examine the relationship between food consumption and disease, we compared the monthly food purchases of the patient group before the disease occurred and the food purchases of the control group. It was determined that the interaction existed at the significance level α=0.1 as a result of testing the effect of two-way ANOVA interaction between the variable dividing the patient group and the control group and the purchased food. In addition, as a result of the post-hoc test, there was an interaction effect between the patient group and control group in fruits, livestock products, aquatic products, seasoned vegetables, confectionery, and fruits and vegetables. Moreover, there was a difference in food consumption patterns between the patient group and the control group.

    Key words :

    더보기

    목차 (Table of Contents)

    • Ⅰ. 서 론 1
    • 1.1 연구배경 및 목적 1
    • 1.2 연구방법 및 범위 2
    • Ⅰ. 서 론 1
    • 1.1 연구배경 및 목적 1
    • 1.2 연구방법 및 범위 2
    • Ⅱ. 이론적 배경 및 관련연구 4
    • 2.1 통계적 방법 4
    • 2.1.1 이원배치 분산분석(Two-way-ANOVA) 4
    • 2.2 군집분석 방법 5
    • 2.2.1 계층적 군집분석 5
    • 2.2.2 비계층적 군집분석 6
    • 2.3 클래스 불균형 데이터의 분류 성능 향상을 위한 방법 7
    • 2.4 머신러닝 분류분석 방법 8
    • 2.4.1 로지스틱 회귀분석 9
    • 2.4.2 의사결정나무 10
    • 2.4.3 XGBoost 12
    • 2.5 관련 연구 13
    • 2.5.1 식품 소비와 대사성 질환 연관성 연구 13
    • 2.5.2 이원변량 분산분석 (Two-way ANOVA) 14
    • Ⅲ. 연구대상 및 방법 16
    • 3.1 연구대상 정의 16
    • 3.1.1 변수 정의 16
    • 3.1.2 데이터 전처리 19
    • 3.2 연구 방법 20
    • 3.2.1 Two-way ANOVA 20
    • 3.2.2 군집분석 20
    • 3.2.3 머신러닝 20
    • Ⅳ. 분석 결과 22
    • 4.1 연구대상자의 인구통계학적 특성 20
    • 4.2 환자군-대조군의 통계적 분석을 통한 특성 비교 24
    • 4.3 식품 소비 특성과 대사성 질환 간 상관 분석 25
    • 4.4 식품 소비 특성의 정의를 위한 군집분석 28
    • 4.4.1 대사성 질환 진단 후 군집 이동 및 특성 비교 31
    • 4.5 식품 소비에 따른 대사성 질환 분류분석 32
    • 4.5.1 SMOTE를 활용한 클래스 균형화 32
    • 4.5.2 로지스틱 회귀를 이용한 분류 모델 33
    • 4.5.3 의사결정나무를 이용한 분류 모델 34
    • 4.5.4 XGBoost를 이용한 분류 모델 36
    • Ⅴ. 결론 및 향후 연구 37
    • 참 고 문 헌 40
    더보기

    분석정보

    View

    상세정보조회

    0

    Usage

    원문다운로드

    0

    대출신청

    0

    복사신청

    0

    EDDS신청

    0

    동일 주제 내 활용도 TOP

    더보기

    주제

    연도별 연구동향

    연도별 활용동향

    연관논문

    연구자 네트워크맵

    공동연구자 (7)

    유사연구자 (20) 활용도상위20명

    이 자료와 함께 이용한 RISS 자료

    나만을 위한 추천자료

    해외이동버튼