信用風險評估模式之建構－應用資料探勘法

CCUR > College of Business > Department of Accounting & Graduate Institute of Accounting > Thesis > Item 987654321/25710

Please use this identifier to cite or link to this item: https://irlib.pccu.edu.tw/handle/987654321/25710

Title:	信用風險評估模式之建構－應用資料探勘法 Credit Risk Model Construction by Data Mining Techniques
Authors:	周澄俊 Chou, Cheng-Jun
Contributors:	會計學系
Keywords:	信用風險維度縮減決策樹 corporate credit risk dimensionality reduction decision tree
Date:	2013-06
Issue Date:	2013-10-22 13:59:06 (UTC+8)
Abstract:	公司信用風險之衡量受到許多研究學者之關注，在近期的相關研究中指出人工智慧之技術優於傳統統計分析之技術，有鑑於此，本研究提出一個新穎的預測模型，此模型整合兩個重要之機制，一為維度縮減，另一個為人工智慧模型的構建，維度的縮減是用來降低維度之影響，資料的維度過高容易使模型預測效果下降，若資料經過維度縮減可以降低模型運算的成本與提高預測準確率，將原始資料經過維度縮減之程序再進一步匯入人工智慧的技術並構建出一個有效的預測模型，在人工智慧技術的採用上以決策樹為本預測模型的主體，經過實證結果的檢驗，本研究所提出的預測模型較佳，決策者可以採用此預警模型作為調整個人資產配置與決策分析的參考依據。 Corporate credit risk assessment has got considerable attention in prior researches, and recent researches have shown that artificial intelligences (AI) reached superior per-formance than traditional statistical techniques. The study proposed an emerging fore-casting mechanism which incorporated dimensionality reduction technique and artificial intelligence. The dimensionality reduction technique is used to alleviate the curse of dimensionality and decrease the computational cost as well as enhance the forecasting performance. The raw data undergone the dimensionality reduction process was fed into decision tree (DT) to construct the forecasting mechanism. The experimental results show that the proposed mechanism outperforms the other statistical models. The deci-sion makers can take the forecasting mechanism as a pre-warning model to modify the investing strategies and adjust personnel wealth.
Appears in Collections:	[Department of Accounting & Graduate Institute of Accounting] Thesis

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	379	View/Open

View Licence

Loading...