查詢結果分析
來源資料
相關文獻
- 資料探索方法在醫學資料庫之評估
- Applying Decision Tree and Neural Network to Raise the Performance of Human Training Quality
- Standard WIP Determination and WIP Balance Control with Time Constraints in Semiconductor Wafer Fabrication
- 應用資料採礦技術分析生物科技新產品之市場區隔:以健康食用油為例
- 信用評等、公司違約率與財務危機預測之探討
- 應用機器學習演算法於乳癌資料分析之研究
- 手寫數字辨識模式之建立--結合遺傳演算法與類神經網路
- Form Segmentation and Component Classification for Clinic Document Image Analysis
- 倒傳遞類神經網路在波浪時序列預報之應用
- 資料挖掘的運用--以決策樹方法分析保送甄試學生入學四技二專報到情形
頁籤選單縮合
題名 | Applying Decision Tree and Neural Network to Raise the Performance of Human Training Quality=應用決策樹與類神經網路以提升人力訓練品質之績效 |
---|---|
作者 | 周永燦; 吳益銓; 林文燦; | 書刊名 | 品質學報 |
卷期 | 22:5 2015.10[民104.10] |
頁次 | 頁383-403 |
專輯 | Artificial Intelligence and Data Mining to Quality Management |
分類號 | 494.386 |
關鍵詞 | 倒傳遞類神經網路; 決策樹; Back-propagation neural network; Decision tree; IIP; K-Means; TTQS; |
語文 | 英文(English) |
中文摘要 | 人力資源是組織中最重要的資產。在全球化的競爭趨勢下,人力資源成為公司主要的核心部門,最根本的工作任務就是建立優秀的人力培訓系統以提高訓練品質。台灣訓練品質系統TTQS為協助國內事業單位、訓練機構針對內部人力教育訓練的執行,提供一套完善的教育訓練品質系統,藉由TTQS系統的導入及運作,並依照TTQS制度之PDDRO(Plan、Design、Do、Review、Outcomes)五環程序標準及19項評核指標,評估企業所有人力訓練方案過程及結果之優劣,使執行後的結果與企業績效成為一個系統性的整體規劃,讓人力教育訓練能更符合事業單位的需要。本研究應用資料探勘技術探討人力訓練品質與績效,以2012年TTQS新版評核資料庫為探勘基礎,尋找TTQS之關鍵評核指標項目。首先應用倒傳遞類神經網路評估TTQS資料庫分類之準確率與績效,其訓練樣本與測試樣本預估準確率都高達95%以上。接著比較決策樹演算法(C5.0,CART,CHAID),選擇準確率較高之演算法來探討TTQS關鍵指標,發現C5.0演算法無論在任何資料分割比例下,皆保有較佳之學習準確率,其測試準確率最高達89.41%。K-Means集群分析法用來驗證比較決策樹演算法的結果。經交叉比對K-Means集群分析與決策樹C5.0演算之結果,本研究找出TTQS之重要關鍵評核指標九項。研究結果可幫助台灣之公司企業導入台灣訓練品質系統,掌握TTQS評核關鍵指標內容,強化人力訓練品質與績效。 |
英文摘要 | Human resources are the most important asset of organization. In the trend of globalization, human resources turned to be the core department of a company. The fundamental job is to establish an excellent human training system and enhance training quality. Taiwan TrainQuali System, TTQS provides businesses and training institutions with the tools to carry out internal human training by offering a sound system of evaluating the education and training. The assessment of how good the processes and outcomes of all the training programs in an enterprise by operating the TTQS and following its PDDRO (Plan, Design, Do, Review, Outcomes) procedural standards can integrate the executive results and the enterprise's performance into a systematic integral plan, making the education and training better meet the business needs. This study applies data mining techniques to explore human training quality of 2012 TTQS new version assessment review database and to find TTQS critical assessment indicators. Back-propagation neural networks first apply to evaluate TTQS database classification accuracy and performance. Training patterns and testing patterns prediction accuracy are greater than 95%. Then the study analyzes and compares decision tree algorithms (C5.0, CART, CHAID), and chooses a higher accuracy rate algorithm to discuss TTQS critical indicators. C5.0 has better accuracy rate under any partition proportions and the highest testing accuracy is 89.41%. K-Means clustering analysis is to identify the critical indicators chosen by the decision tree. Through cross comparison with decision tree and K-Means results, this study identifies 9 important critical indicators of TTQS assessment to help enterprise in Taiwan to introduce TTQS, to grasp TTQS assessment critical indicators, and to enhance the quality of human training and performance. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。