應用資料探勘技術於醫美機構評量之研究

ASIA unversity > 管理學院 > 經營管理學系 > 博碩士論文 > Item 310904400/113235

Please use this identifier to cite or link to this item: http://asiair.asia.edu.tw/ir/handle/310904400/113235

Title:	應用資料探勘技術於醫美機構評量之研究 A Study of Applying Data Mining Approach to the Evaluation of Aesthetic Medicine Institutes
Authors:	林明怡 LIN, MING-YI
Contributors:	經營管理學系
Keywords:	美容醫學;資料探勘;分類法則;集成法 Aesthetic Medicine;Data Mining;Activity Based Classification;Ensemble Learning
Date:	2022
Issue Date:	2022-06-13 09:12:00 (UTC+0)
Publisher:	亞洲大學
Abstract:	中國大陸美容醫學市場增速高於全球市場，雖然近年增速有所放緩，但從滲透率看，中國大陸市場的增長空間仍較為廣闊。根據艾瑞咨詢的數據顯示，2015-2019年，中國大陸美容醫學需求增速超過全球，中國大陸美容醫學市場規模從2015年的人民幣648億元人民幣一路攀升至2019年的1,769億元人民幣，年復合增長率達28.7%。但受到新冠疫情以及市場供大於求的影響，2018年開始中國大陸美容醫學行業的增速放緩。2020年-2023年的年復合增長率預計降為15.2%。但艾瑞咨詢同時預測，預計行業經過3-5年的行業自我調整和變革後，市場將逐步回暖，2023年中國大陸美容醫學市場規模將達到人民幣3,115億元人民幣。本研究的主要目的是探討某美容醫學產業產品在中國大陸美容醫學機構銷售狀況，並對其各機構進行星級分類。在本研究中我們採用基於決策樹的資料探勘技術來探索該公司在中國大陸銷售的美容醫學機構等級的分類規則。此外，還應用通過決策樹算法增強集成法構建的多學習器模型。數值結果表明，使用多個模型提高了分類精度。特別是從數據挖掘方法中提取的規則可以開發為計算機模型，用於分類美容醫學機構評等優劣之專家系統。 The growth rate of the medical beauty market in mainland China is higher than that of the global market. Although the growth rate has slowed down in recent years, in terms of penetration rate, the growth potential of the market in China is still quite high. According to data from iResearch, from 2015 to 2019, the growth rate of demand for aesthetic medicine in mainland China exceeded that of the world. The market size of aesthetic medicine in mainland China has climbed from RMB 64.8 billion in 2015 to RMB 176.9 billion in 2019, with a compound annual growth rate of 28.7%. However, due to the impact of the new crown epidemic and the oversupply of the market, the growth rate of the medical beauty industry in mainland China has slowed down since 2018. The compound annual growth rate from 2020 to 2023 is expected to drop to 15.2%. However, iResearch also predicts that the market is expected to gradually pick up after 3-5 years of industry self-adjustment and reform. In 2023, the market size of aesthetic medicine in mainland China will reach RMB 311.5 billion. The main purpose of this study is to explore the sales status of a cosmetic medicine company's products in Chinese cosmetic medicine institutions, and to classify its various institutions by star rating. In this study, we use decision tree-based data mining technology to explore the company's classification of star ratings according to the sales status of various aesthetic medicine institutions. In addition, a multi-learner model constructed by augmented ensemble of decision tree algorithms is applied. Numerical results show that using multiple models improves classification accuracy. In particular, the rules extracted from the data mining method can be developed into a unique computer model, just as an expert system can be constructed to rate the beauty medical institutions.
Appears in Collections:	[經營管理學系 ] 博碩士論文

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	93	View/Open

View Licence

Loading...