본문 바로가기
KERD
논문KCI 등재

보건정보 자료를 이용한 나무구조회귀모형의 시각화에 대한 고찰

Visualization with Various Tree-based Regression Results in Health Information Data

이은경|조현선
한국보건정보통계학회 (The Korean Society of Health informatics statistics)|2020.02|보건정보통계학회지 (Journal of Health Informatics and Statistics)|제45권 제1호|pp. 16-29|DOI https://doi.org/10.21032/jhis.2020.45.1.16

1

조회

0

다운로드

0

인용

0

참고문헌

키워드

Data visualizationRegression analysisBaggingTree-based regression modelBoosting

초록

Objectives: Among machine learning techniques, a tree-based regression model is widely used as easy to interpret and easy to use results. Methods: In this study, we examine the characteristics of regression models using various tree structures implemented in R, and apply them to public health data for visual representation and analysis to enhance understanding of the models and data. Results: We also look at the random forest, gradient boosting model, and xgboost that incorporate the results of various tree-structured model estimates using bagging and boosting, and visualize these estimation processes to compare the results of the model estimates. It also compares the performance of various tree-structured models using various public health data. It visually examines that the performance of the tree-structured models may vary depending on the characteristics of the data. Conclusions: Through this study, we apply the tree-structured models to public health data to enhance understanding of the data and to help identify and visualize the mechanism of the data.

Abstract

Objectives: Among machine learning techniques, a tree-based regression model is widely used as easy to interpret and easy to use results. Methods: In this study, we examine the characteristics of regression models using various tree structures implemented in R, and apply them to public health data for visual representation and analysis to enhance understanding of the models and data. Results: We also look at the random forest, gradient boosting model, and xgboost that incorporate the results of various tree-structured model estimates using bagging and boosting, and visualize these estimation processes to compare the results of the model estimates. It also compares the performance of various tree-structured models using various public health data. It visually examines that the performance of the tree-structured models may vary depending on the characteristics of the data. Conclusions: Through this study, we apply the tree-structured models to public health data to enhance understanding of the data and to help identify and visualize the mechanism of the data.

참고문헌 (36)

  1. Malehi AS, "Enhanced Expert Systems" (2019), Intech Open
  2. Fonarow GC, "Risk stratification for in-hospital mortality in acutely decompensated heart failure : classification and regression tree analysis", JAMA, 293(5), pp.572~580 (2005)[DOI]
  3. Cheng Z, "Evaluation of classification and regression tree(CART)model in weight loss prediction following head and neck cancer radiation therapy", Adv Radiat Oncol, 3(3), pp.346~355 (2017)[DOI]
  4. Karafin MS, "The impact of recipient factors on the lower than expected hemoglobin increment in transfused outpatients with hematologic diseases", Transfusion, 59(8), pp.2544~2550 (2019)[DOI]
  5. Loh WY, "A regression tree approach to identifying subgroups with differential treatment effects", Stat Med, 34(11), pp.1818~1833 (2015)[DOI]
  6. Loh WY, "Identification of subgroups with differential treatment effects for longitudinal and multiresponse variables", Stat Med, 35(26), pp.4837~4855 (2016)[DOI]
  7. Dusseldorp E, "The regression trunk approach to discover treatment covariate interaction", Psychometrika, 69(3), pp.355~374 (2004)[DOI]
  8. Dusseldorp E, "Combining an additive and tree-based regression model simultaneously : STIMA", J Comput Graph Stat, 19(3), pp.514~530 (2010)
  9. Morgan JN, "Problems in the analysis of survey data, and a proposal", J Am Stat Assoc, 58(302), pp.415~434 (1963)
  10. Einhorn HJ, "Alchemy in the behavioral sciences", Public Opin Q, 36(3), pp.367~378 (1972)
  11. Doyle P, "The use of automatic interaction detector and similar search procedures", J Oper Res Soc, 24(3), pp.465~467 (1973)
  12. Breiman L, "Classification and regression trees", pp.1984, Chapman & Hall
  13. Quinlan JR, "Learning with continuous classes", 5th Australian Joint Conference on Artificial Intelligence, pp.343~348 (1992), World Scientific
  14. Wang Y, "Induction of model trees for predicting continuous classes", Poster papers of the 9th European Conference on Machine Learning (1996), Springer
  15. Torgo L, "Functional models for regression tree leaves", ICML, pp.385~393 (1997)
  16. Chaudhuri P, "Piecewise-polynomial regression trees", Stat Sinica, 4(1), pp.143~167 (1994)
  17. Loh WY, "Regression tress with unbiased variable selection and interaction detection", Stat Sinica, 12(2), pp.361~386 (2002)
  18. Hothorn T, "Unbiased recursive partitioning: a conditional inference framework", J Comput Graph Stat, 15(3), pp.651~674 (2006)[DOI]
  19. Kim H, "Visualizable and interpretable regression models with good prediction power", IIE Transactions, 39(6), pp.565~579 (2007)[DOI]
  20. Loh WY, "Handbook of Data Visualization", pp.447~469 (2008), Springer
  21. Garcia AL, "Improved prediction of body fat by measuring skinfold thickness, circumferences, and bone breadths", Obes Res, 13(3), pp.626~634 (2005)[DOI]
  22. Hothorn T, "ctree: Conditional inference trees", The Comprehensive R Archive Network, pp.1~34 (2015)
  23. Breiman L, "Bagging predictors", Mach Learn, 24(2), pp.123~140 (1996)[DOI]
  24. Friedman JH, "Additive logistic regression: a statistical view of boosting", Ann Stat, 28(2), pp.337~374 (2000)
  25. Breiman L, "Random forests", Mach Learn, 45(1), pp.5~32 (2001)[DOI]
  26. Friedman JH, "Greedy function approximation : a gradient boosting machine", Ann Stat, 29(5), pp.1189~1232 (2001)
  27. Friedman JH, "Stochastic gradient boosting", Comput Stat Data Anal, 38(4), pp.367~378 (2002)[DOI]
  28. Chen T, "XGBoost: a scalable tree boosting system", 22nd SIGKDD Conference on Knowledge Discovery and Data Mining (2016)
  29. Ridgeway G, "Generalized boosted models: a guide to the gbm package", ~ (2007)
  30. Chen T, "Xgboost: extreme gradient boosting", R package version 0.4-2, ~ (2015)
  31. Chipman HA, "BART : Bayesian additive regression trees", Ann Appl Stat, 4(1), pp.266~298 (2010)[DOI]
  32. Kapelner A, "bartMachine : machine learning with Bayesian additive regression Trees", J Stat Softw, 70(4), pp.1~40 (2016)[DOI]
  33. Biecek P, "DALEX : explainers for complex predictive models in R", J Mach Learn Res, 19(1), pp.3245~3249 (2018)
  34. Gosiewska A, "auditor: an R Package for Model-Agnostic Visual Validation and Diagnostic", arXiv 2018; preprint arXiv:1809.07763, ~
  35. Kenkel DS, "The effect of physician advice on alcohol consumption : count regression with an endogenous treatment effect", J Appl Econ, 16(2), pp.165~184 (2001)
  36. Hilbe JM, "Negative binomial regression" (2011), Cambridge University Press

관련 논문AI추천

보건의료 빅데이터의 시각화 방법들강대룡, 김성화, 임지혜변수 중요도를 이용한 설명 가능한 인공지능 기법의 시각화에 대한 고찰과 보건정보 자료에의 응용박정훈, 이영섭, 임창원, 정혜린한국보건정보통계학회 · 보건정보통계학회지2020 · 제45권 제4호임상자료를 이용한 나무구조 분류모형의 성능 비교신혜정, 이윤동, 이은경한국보건정보통계학회 · 보건정보통계학회지2016 · 제41권 제1호R을 이용한 결과 변수에 따른 예측 모형과 시각화 - 회귀분석을 중심으로양주연, 이혜선, 전소영국민건강검진데이터기반혈색소(헤모글로빈) 예측모델링정대원, 황욱연한국품질경영학회 · 품질경영학회지2024 · 제52권 제4호기계학습모형의 시각화허명회국가데이터연구원 · 통계연구2015 · 제20권 제2호

이 간행물의 다른 논문 (11건)

Medical Neglect by Parents and Child Health at Clinical Sites최소영, 배화옥, 서지현pp. 1-7치과 내원 흡연자의 전자담배 사용 및 사용의도 관련요인류소연, 최성우, 한미아, 김소영, 박종pp. 8-15간호대학생의 공감능력과 인권감수성이 다문화수용성에 미치는 영향노윤구, 안수라pp. 30-36일개 종합병원 간호사의 멘토링 기능이 심리적 안녕감에 미치는 영향박종, 최성우, 류소연, 김은숙pp. 37-43퇴원손상심층조사 자료의 효율적 활용을 위한 데이터베이스 구축에 대한 연구김성수pp. 44-51
원문 보기