Data Mining Applied to One multinational electronic enterprise analysis of the implementation educational training courses

• Main Authors: Yuan-Ching Wu, 吳苑菁
• Other Authors: Wen-Tsann Lin
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/83592937001106476068

碩士 === 國立勤益科技大學 === 工業工程與管理系 === 101 === The enterprises in Taiwan are unable to realize their education and training because they have resource limitation to make the internal training happen. To enhance the quality of human resources, how to make education more efficient planning of training courses is certainly an important issue for enterprises. Knowledge economy had been growing rapidly in recent years. In the trend of globalization, human resources turned to be the core department of companies. The fundamental job is to establish quality human resources environment and enhance human resources quality. The training framework is designed in accordance with the spirit of Taiwan TrainQuali System (TTQS). The training structure in the confirmation P (Plan), D (Design) two steps, the results of the following up will be more easily and smoothly. This research is based on the TTQS database by BEVT that enterprises outside the control limits are down in P and D dimensions, and it should do some reinforcement for the gap. In this study, an electronics company, for example, through data mining techniques from the database to identify the implicit or not obvious information that can help to plan and design future education training courses. This study uses data mining developed the theoretical two-step clustering method (SOM and K-means). Collecting an electronic company both Taiwan factory and the mainland factory personnel education and training information performed mining and analysis. Calculated by the results of these two algorithms, it develops training courses for future reference to enhance the effectiveness of company personnel involved in training. Decision tree analysis to explore the importance differences of database between mainland factory and Taiwan factory, Finally, Back-propagation neural network combined to develop an education and training prediction model. The results of this study can be used as a basis for future training arrangements, and how to plan for courses that have maximum benefit. To identify the key success factors of education and training, other companies can be developed as a future strategy for education and training of reference and to enhance competitiveness and innovation.