Regional Human Resource Management And Decision Platform Based On Data Mining

With the rapid development of computer technology and Internet technology,human resource management has been modernized and technicalized. However, the traditional statistical methods can not meet the requirements of the government and enterprises on the introduction of talent, talent training and talent reserve. Therefore, using the data mining technology to find hidden rules in human resources management system(HRMS) and providing decision support for the government and enterprises has become an urgent concern of the government and enterprises.This paper starts from the study of human resources management system model framework and data warehouse design. Combining integration idea with open-source Weka interface, a combining classifier model is proposed based on multiple decision tree algorithms. The data mining model based on rule classification and ant colony algorithm is improved, which has been applied to the data mining of HRMS which obtained the ideal classification effect and effective decision support. The innovation work of this paper mainly includes:First of all, the design method of business architecture of human resource management system is proposed, meanwhile,the subject-based human resource data warehouse and the associated system structure are designed via the combination of data warehouse theory and HRMS.Secondly, the weighted voting method is adopted to develop a new combining classifier based on the typical decision tree algorithms in Weka.After simulating the data in HRMS by the new classifier, the results illustrate that classification accuracy of the combining classifier is superior to that of a single one.Thirdly, in order to achieve a balance between convergence speed and the global searching capability of the Ant-Miner algorithm, this paper adjusts the probability of the deterministic selection and the volatility coefficient dynamically and adaptively. The effectiveness of the improved algorithm is verified in the data set of UCI public database. Furthmore,the algorithm also obtained effective classification rules in the human resource management system.