A Research On Association Rule Mining Method For Colleges' Multi-source Heterogeneous Data

Information technology enters the era of big data,and the rapid development of data generation,transmission and storage technologies has led to the massive growth of data and the rapid expansion of the search space,it also has led to a large increase in heterogeneous data and an increasingly serious problem of exchange between heterogeneous data.In the process of managing the use and storage of massive data,it has inspired people to conduct research on the analysis and mining of a large amount of existing data.How to extract valuable indirect expression information from large-scale massive data,the data mining field has carried out a lot of in-depth research,how to discover the relationship between transactions from heterogeneous massive data is a powerful challenge.In the process of information construction,colleges and universities have accumulated a large amount of heterogeneous data.Using data mining technology to mine learning data and teaching data can assist teaching management and further improve the comprehensive strength of the school.The demand for correlation analysis and mining of big data in colleges and universities is getting stronger and stronger,and the big data mining process of colleges and universities has gradually begun.When the user mines the association rules for a large amount of data,the algorithm is ineffective due to the increase of the data dimension and the data size.When the association rule mining algorithm is applied to a large amount of heterogeneous data,the effectiveness is limited by the memory.On the basis of in-depth study of university domain metadata integration and full understanding of association rule mining algorithms,integrating the metadata of colleges and universities,improving the association rules mining algorithm based on the integration of college metadata,improving the execution efficiency of the algorithm in specific fields,and improving the quality of rule generation,this paper proposes multi-source heterogeneity for colleges and universities.A method for mining association rules for data.The method includes a metadata integration method for multi-source heterogeneous data environment in colleges and a association rule mining algorithm based on metadata integration.The metadata integration method for multi-source heterogeneous environment in colleges is divided into four parts: firstly,the local ontology is constructed,and then the domain ontology is constructed on the basis of the mapping between local ontology and local ontology.The local metadata is extracted,and finally the global metadata is integrated under the guidance of the university domain ontology.The integrated college metadata is introduced into the association rules mining process,and the frequent pattern mining links are improved to form an association rule mining algorithm based on metadata integration.First,the data is sampled and rules are generated under the guidance of integrated metadata.Then,the results of the sample generation rules are used to calculate the attribute relevance for data block operations.Finally,by constructing a local frequent pattern tree and constructing a global frequent pattern.The tree and association rules are generated to complete the algorithm.The algorithm proposed in this paper is experimentally verified.Experiments show that the algorithm is better than mainstream algorithms in mining large-scale data.In summary,the main contents of this paper include: the research background,research status and research significance,as well as related literature;This paper proposes a metadata integration method for multi-source heterogeneous data environment in colleges and universities,and further proposes a frequent pattern mining algorithm based on metadata integration.This paper compares and analyzes the association rules mining methods for multi-dimensional heterogeneous data in colleges and universities,and further evaluates the experimental results.