| With the increase in the number of graduates of our school, employment and employment direction is more and more heavy, the research and development of employment information management system, is very necessary to enhance employment level and employment direction of the graduates in our college. How to use data mining technology to obtain employment guidance information, it is particularly important to improve the working efficiency, this paper mainly from the following three aspects:Aiming at a great deal of data existing employment information system, through the analysis of sampling data preprocessing, and obtain key information, and employment are closely related to the graduates information such as:comprehensive achievement, English level, the ability of computer, personal character orientation and a series of data.A lot of problems in the candidate set and repeat scanning database generated for the Apriori algorithm, this paper use APriori multidimensional frequent items set algorithm based on the transaction database, through mapped to a Boolean matrix, using method of layer by layer increased to dynamically allocated memory reuse vector for "and" operation of finding frequent itemsets. Algorithm for generating candidate set to get a large reduction, and is applied to the university employment management system, which effectively shorten the scan time and save the memory cost, improve work efficiency.Based on above mining association rules, the paper uses the C4.5 algorithm to construct the decision tree of employment, which graduate students be classified according to the corresponding computer ability, English level, comprehensive performance, political affiliation, and other personality orientation decision attribute, in order to study the influence of graduates' employment categories and direction, thereby to improve the employment rate, level improvement of the existing training mechanism, etc. and to provide decision support for manager. |
PDF Full Text Request