應用分群演算法尋找影響企業辦理訓練品質關鍵因素之研究

Research of a Clustering Algorithm for Enterprise on key factors affecting the quality of training

張鴻文(Hung-Wen Chang)

指導教授 ： 陳同孝 ； 陳民枝

國立臺中科技大學/資訊與流通學院/資訊工程系碩士班/碩士(2012年)

摘要

在進入知識經濟的時代後，人力資源已成為企業最重要的生產要素之一，故人才培訓亦成為企業發展的重要工作。但是國內多數企業對於教育訓練並無明確的評估標準及分析機制，以致辦理成效不一。所以，政府為了提升企業辦訓能力，進而提升國家競爭力，制定一套訓練品質評核系統(Taiwan TrainQuali System, TTQS)，並運用訓練品質計分卡對企業與訓練單位進行評核，除了可以協助企業與訓練單位自我檢視外辦訓能力與績效外，也可作為政府評量辦訓品質的辨識工具，經評核符合補助門檻的企業均給予適當之經費補助。
企業辦理員工教育訓練是否能提升企業的競爭力，而企業的產業屬性與訓練品質計分卡各項指標之評核結果又是否有關聯性。本研究將探討各項訓練品質計分卡指標評核結果與企業訓練成效之關聯性。本研究使用分群演算法，從訓練品質評核系統資料庫中，進行資料探勘與分析，找尋影響企業訓練成效之關鍵因素。
實驗結果顯示在不同產業屬性下，訓練品質計分卡中的設計構面及執行構面，分別與成果構面有相當高的關聯性，是影響企業訓練成效最重要的關鍵因素，訓練課程設計得宜，進而完成訓練的執行，可提高最後的成果展現。實驗結果可做為未來企業擬定教育訓練策略之參考，達成企業提升辦訓品質之目標。

關鍵字

資料探勘 ； 分群演算法 ； 人力資源管理 ； 訓練品質評核系統

並列摘要

In knowledge economy era, human resource has become one of the most important manufacturing factors in the enterprise. Therefore, human resource incubation has become an important job for corporate development. However, for most of the domestic enterprises, there is no clear assessment standard and analytical mechanism for educational training, the implementation result is thus not good. Hence, in order to enhance enterprise’s capability in implementing training and in turn enhance the country’s competitiveness, the government has prepared a set of training quality audit system, namely, Taiwan TrainQuali System (TTQS), meanwhile, training quality scorecard has been used to evaluate the enterprise and training department, hence, in addition to helping enterprise and training department to self-inspect the training implementing capability and performance, it can also be used as a recognition tool for the government to evaluate the training implementing quality, in other words, enterprise evaluated to meet the support standard will be given with appropriate financial support.
This study is going to investigate whether the implementation of employee’s educational training by an enterprise can enhance the corporate competitiveness or not, and whether correlation exists between the industry attribute or industry type of an enterprise and the audit result of the index of the training quality scorecard. Moreover, this study is also going to investigate the correlation between the audit result of the index of training quality scorecard and the training performance of an enterprise. In this study, clustering algorithm is used, and data mining and analysis technique will be done on the database of the training quality audit system so as to find out key factors affecting the training performance of an enterprise.
The experimental results show that the properties of the different industries, training quality scorecard design dimensions and implementation aspects were very high correlation with the dimensions of the outcome, the most important key factors influence the effectiveness of corporate training, training design properly, and then completed the implementation of training, can improve the final results show. The experimental results can be used as future enterprises to develop education and training strategy of reference to reach the target enterprises to enhance their quality do training.

並列關鍵字

Data mining ； Human resources management ； Clustering algorithm ； Taiwan TrainQuali System(TTQS)

參考文獻

[12]S.H. Tsai, Strategic Applications of Employees' performance: Perspective of High Performance Public Organization, YanKaoShuangYueKan, Vol31,No.2,pp.40-54, 2007.

[4]T.S. Chen, C.C. Lin, Y.H. Chiu, H.L. Lin and R.C. Chen, A New Binary Classifier: Clustering-Launched Classification, Lecture Notes in Artificial Intelligence, pp.278-283, 2006.

[5]J. Han and K. Micheline, Data Mining: concepts and techniques, Morgan Kaufmann, 2001.

[8]ISO 10015，Quality management - Guidelines for training.

[9]J. B. MacQueen, "Some Methods for classification and Analysis of Multivariate Observations", Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, University of California Press, 1:281-297, 1967.