Rapidminer大數據挖掘於成人健康檢查之慢性病預防

Apply Big Data Mining with Rapidminer to Adult Health Examination for Prophylaxis of Chronic Illness

楊峒禹(Tong-Yu Yang)

指導教授 ： 洪士程

朝陽科技大學/資訊學院/資訊工程系/碩士(2017年)

摘要

由於人口老化及人口結構的改變，高齡化的相關社會議題，漸漸受到大家的關注及重視。人口老化帶來醫療費用高漲、慢性病、身體功能退化、殘障等問題成為家庭、社會、國家的沉重負擔。因此，藉由健康檢查來達到早期發現早期治療等目標是根本之道。成人健檢資料中，潛藏著一些未曾探索的重要資訊與知識。大數據挖掘方法可以做資料類型的發覺與萃取，而對一資料組做最佳的詮釋。本篇論文用Rapidminer大數據挖掘軟體，去從健康檢查資料中找出一些潛藏的重要資訊。本研究是以1158位成人的健康檢查資料為資料庫，資料分析分成二大階段，第一階段使用SPSS統計軟體找出危險因素，第二階段使用Rapidminer大數據挖掘軟體對這些資料建立決策樹。決策樹是一個功能強大的預測模型，他可以生成規則容易理解的模型，並且可以看出屬性與類別之間的相關性。藉由決策樹模型可以從中挖掘出隱含規則，將這些分析結果提供給醫生作為輔助參考，用來提高診斷的正確性以及診斷速度。

關鍵字

健康檢查 ； 慢性疾病 ； 決策樹 ； Rapidminer ； SPSS ； WEKA

並列摘要

Due to increased number of aging population and population structure changes, people are concerned about aging-related problems that put great burdens and make enormous impacts on the family, society, and nation. Aging population brings along chronic disease, deterioration of boding function and disabilities resulting in increase medical expenditure and heavy burden to the family, society and the nation as a whole. Regular medical checkup and early treatment is the most effective solution to this problem. Important but non-intuitive information and knowledge in the adult health check-up database can be found, extracted and organized using big data mining. This thesis applied big data mining techniques with Rapidminer to enhance understanding of abnormal item combinations in adult health check-up data. Researchers collected a total of 1,158 health examination records. Data analysis consisted of the two parts of general statistics by SPSS and big data mining with Rapidminer. General statistics was a pre-process used to clarify and organize data in preparation for big data mining. A decision tree is proposed to solve the classification problem with large number of classes and continuous attributes. This research uses the decision tree to explore abnormal item combination on community health screening services for the elderly. Besides, related factors of the information and knowledge are also discussed. Suggestions may serve as a useful reference for doctor to improve the correct diagnosis of a chronic illness.

並列關鍵字

Health examination ； Chronic illness ； Decision tree ； Rapidminer ； SPSS ； WEKA

參考文獻

[1] 張士瑋，應用資料探勘技術於成人健康檢查之慢性病預防，碩士論文，朝陽科技大學資訊工程系，臺中，2014。

[2] Bartosz Wojciuk, Marek Myslak, Krzysztof Pabisiak, Kazimierz Ciechanowski, Stefania Giedrys-Kalemba1, "Epidemiology of infections in kidney transplant recipients – data miner’s approach," Transplant International, 2015, ISSN 0934-0874.

[3] Goudy Steven, Jiramongolchai Pawina, Chinnadurai Sivakumar, " Logistic regression analysis of Pierre Robin sequence patients requiring surgical intervention," Laryngoscope, April 2017, vol. 127, issue 4, pp. 945-949.

[4] Meurer William J., Tolles Juliana, "Logistic Regression Diagnostics Understanding How Well a Model Predicts Outcomes," JAMA, March 2017, vol. 317, issue 10, pp. 1068-1069.

[5] Meng Lei, Ford Trent, Guo Ying, "Logistic regression analysis of drought persistence in East China," International Journal of Climatology, March 2017, vol. 37, issue 3, pp. 1444-1455.