none

• Main Authors: Yu-Wen Chen, 陳俞文
• Other Authors: Nien-Chi Liu
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/16799040000057944131

碩士 === 國立中央大學 === 人力資源管理研究所 === 102 === In recent years, enterprises pay lots of attention on employees’ benefits and expect to satisfy employees’ requirement, improve the job satisfaction, job involvement or increase the organization commitment; in addition, the benefits system can increase the organization’s performance. Companies need to know what employees need about their benefits. By reviewing literatures, we can know that the characteristic is one of the causes of the differentiation of benefits’ need. If companies can diagnose their employees’ characteristics and figure out their differentiation, this diagnose can help construct a best model fits the organization. In the past literatures, scholars always use the multiple linear regression to examine the relationship between the satisfaction of benefits and the employees’ attitude; however, this relationship might be non-linear. This study uses the systematical data mining technique to investigate the case company’s benefits measures and uses the neural network to analyze the fits between the benefits system and employees’ satisfaction towards benefits, job satisfaction, job involvement and organizational commitment. By the neural network analysis, we can analyze the relationship between the benefits system and satisfaction; by the importance analysis, we can figure out the most sensitive benefit’s measures. Those analyses can provide enterprises as a consultation. The results find that the neural network analysis can construct a prediction model and the whole prediction rate can achieve above 75%. By the importance analysis, this study examines out the most effective benefits measures, and by the characteristics such as gender and marriage analysis, we can find employees have different preference toward the benefits measures. In conclusion, this study expects to provide the neural network analysis as a decision making tools, and hope there will be more data mining techniques to construct better outcomes.