Data Mining Based On Dynamic Partition And The Applications In Teaching Affairs Management System

With the rapid progress of database and network technique, data is undergoing explosive growing. People are trying to find information behind the data. Data mining is brought forth under this condition.Recently, many teaching-aid management tools are designed and developed. These tools provide many useful means for users, such as browsing, searching, and so on. But it is not enough for statist and analyzer. Users need to analyse those data manually to get valuable information from them. It is the topic of this thesis that how to get information and potential knowledge through these data mining. Based on the students' score database, relationships among the courses can be extracted, which is helpful to arrange the teaching plan.As one of the main pattems in the field of data mining, association rules are used to determine the relationships among the attributes or objects, to find out valuable dependencies among the fields. The efficiency of mining frequent item sets is the key problem in association roles generating.The arithmetics of association roles and frequent item sets are discussed in this thesis. The author is enlightened by the shopping basket analysis and uses the thinking of the mining of association rules to mine the students' score database. The author collected and settled the data of score, presented a method of dynamic partition data, and designed arithmetic of data transform. Based on the foregoing works, high quality data sets are generated. We use arithmetics of FP to mine the data sets and obtain the frequent item sets. A new arithmetic of association rules was presented and implemented. We obtain some association roles and analyse these rules one by one and obtain many useful conclusions.In the latter research, we will continue to research frequent item sets and improve the mining efficency.