Study On The Early Warning Problem For Emplyee Turnover Of IT Industry

Since the Reform and Opening and with the rapid development of China's IT industry, the competition between enterprises is increasingly fierce. IT enterprise produces mainly knowledge,so it has strong dependence on human resources.But in the changing market environment,the crisis of employee turnover will occur at any time,especially IT industry holds high levle of employee turnover rate rencetly. IT enterprise can set up an early warning system to reduce a series of losscaused by employee turnover and enhance its core competitiveness.This paper based on the theory of employee turnover,early warning, support vector machine and so on.The survey questionnaire was designed for the main factors of employee turnover and examine the quality of the questionnaire through the analysis of reliability and validity.Then I use the analysis of factor to extract indicators about the model of employee turnover.Tn this paper,the turnover intentions of employee are dentified by support vector machine and the best parameters are obtained by the method of PSO and grid.Then we can found that selecting RBF as the kernel function of classifier can improve the accuracy of classification.The early warning model of employee turnover can identify the intentions of employee's turnover effectively.The Signs of employee turnover can be detected timely by the early warning model of employee turnover.Some suggestions have beeb given to retain talent employee,such as making reasonable salary welfare policy, recruiting talents for the enterprise itself,building a job rotation system, improving the system performance evaluation, perfecting the ladder of professional development, increasing the switching cost of employee turnover and so on. Hopefully this papaer can help to provide effective information for the HR of IT enterprises.