Data Starage Design And Data Mining

The current society is the era of big data and cloud computing.The sorting of data and the information it contains are becoming more and more important.Relatively mature system software,with different disadvantages,has appeared in the aspects of both the storage of data and the association rules mining among data.In this paper,through the research of data storage and data mining,the author has not only established an optimized small data storage database,but also improved the Apriori algorithm in data mining.Its innovative researches are mainly reflected in the following aspects.First,a database is a warehouse built on computer storage devices and organizes,stores,and manages data according to data structures.SQL Server database language is a standard computer language whose function is to access and process databases;C#programming language is an object-oriented high-level programming language that runs on the.NET Framework.After systematically researching the SQL Server database language and C#programming language,the author creatively combines the two languages to construct a small database system that is convenient for statistical data operations.The biggest highlight of this system is its versatility,which can be used for storage of other data as long as appropriate modifications are made.And the practicality of the database system is verified based on the specific case of class fee management system.Second,the Apriori algorithm is a classic association rule mining algorithm in data mining.Its biggest drawback is that it scans the transaction database once every cycle,which not only takes up a lot of space,but also wastes lots of time.After analyzing the Apriori association rules algorithm,the binary association rule algorithm and the matrix association rule algorithm,a method is discovered to improve the Apriori algorithm.This algorithm is an association rule mining algorithm based on frequent itemsets matrix whose biggest feature is the use of frequent itemsets matrix to store transaction data and simplify the data processing,so that it only needs to scan the transaction database once,reducing the storage space and speed up the operation.In the end,the paper has a brief summary and proposes further research directions.