Based On The Quantitative Association Rules Mining Employment Analysis System

In recent years, the employment problems of university graduates in our country appeared unceasingly, the reason is not only the economy reform, enlarging the scale of enrollment, the lack of education resource, and the imperfect employment model, but also the unsuited model of higher education. It is urgent that we deepens the higher education reform, and ameliorates the education model of university. The universities and colleges needs to face the society and train the application students.In this article we attempts to discover the associative relations between the education attributes and the employment attributes of graduates and find the type of application person which the society needs through the data mining technology. So by providing the guidance or the data support to the policy-makers, we will improve the existing higher education model.Association rules mining is one of important matters of data mining. Apriori algorithm is advanced by Agrawal and the others in 1993. At present, association rules technology has been applied to business, telecommunication, finance, agriculture, medical treatment and so on. It has brought a good effect.First the purpose of Apriori algorithm is analyzing the relation of items (attributes) in transaction database. The value of each item is Boolean. But the database of education information includes a lot of quantitative attributes. Later, the prototype of Apriori algorithm was improved and extended by the investigators in order to adapt the different needs of data mining. These improvements include the Partial Completeness.The data tables we processed contain both quantitative attributes and categorical attributes. in this paper we first introduce the measures of partial completeness which quantify the information lost due to partitioning the continual discrete value to intervals. A direct application of this technique can generate too many similar rules. We tackle this problem by using a "greater-than-expected-value" interest measure to identify the interesting rules in the output. This interest measure looks at both generalizations andspecializations of the rule to identify the interesting rules.The increasing of the number of items, which is due to the partition, enlarges the searching coverage and decrease the speed. Therefore we design a measure to represent the both the data set and the items by arrays of bits. We can do the Join and Prune by using bits operation (and, or, xor). This method will speedup the searching by decreasing the data storage space and avoiding frequent page exchange.According to the principle above, we have designed an employment analysis system based on mining quantitative association rules. We also have provided the structure of the system, the function and the detailed design of each module, and the database design. And we developed and implemented the majority of functions. This system has passed an exam and obtained an ideal result.