Multiple Association Rules Mining In Human Resources Based On Matrix

Data mining is at the foremost position of information science and has been successfully applied in many areas. With finding the potential information hidden in huge data, it may forecast some developmental tendency, and the discovered information greatly improves the capability of decision support and making. Association rule, which is one of the most important directions of data mining research, has aroused people's common attention and been widely used gradually. The main goal of association rules mining is to find hidden, interesting and useful rules between attributes in large dataset. Association rules can be classified as boolean association rules and multiple association rules by attributes, the latter one is the foundation of the former one and becomes the focus of KDD. Frequent itemset mining is the key point during association rules mining and we can improve its efficiency by using matrix.Human Resource is a term with which many organizations describe the combination of traditionally administrative personnel functions with performance management, employee relations and resource planning. College Human Resources include, the teachers and researchers, managers and service staff. College Human Resources have the basic characters of human resource and take the task of training the advanced innovative talent.The College Human Resources Management function includes a variety of activities, and the key among them is deciding what staffing needs you have and whether to use independent contractors or hire employees to fill these needs, recruiting and training the best employees, ensuring they are high performers, dealing with performance issues, and ensuring your personnel and management practices conform to various regulations. Activities also include managing your approach to employee benefits and compensation, employee records and personnel policies.The main contributions are as follows:(1) Proposing valid method for data pre-processing, including turning original database into multiple database, deleting the irrelative attributes, providing clean. accurate and targeted data for association rules mining algorithm. Reducing the data process, improve the algorithm efficiency and level of KDD.(2) Proposing the efficient multiple association rules mining algorithm based on matrix, including FAFS algorithm for finding frequent itemset and Gar algorithm for finding association rules in frequent itemset. Prove its high efficiency and validity by tests.(3) Applying the efficient multiple association rules mining algorithm into the College Cadres' Electronical Archives Management Information System for managing college human resources. The successful running of the system validates the feasibility of the algorithm.