Research Of The Association Rules Data Mining Based On Semi-structured Data

The data store in a large database is usually vast, and the work ofdata analysis becomes more and more hard. Data mining technique wasexplored in order to find the useful data efficiently in a database.Semi-structured data XML documents are growing in amount for thesake of the growing demands of data interchanging. And these documentsare feasibly stored in the native XML database. According, it is a point toresearch about association rule mining in native XML databases. Focusedon the difficulties in applications of association rule mining, analyzed thetechnique characteristics of the XML language, this paper proposes amodel based on semi-structured data for association rules mining. Thismodel taken the advantages of semi-structure data source inself-describing, data sharing, and flexibility.Now, many researchers study the relevant technique in data mining,and there is a popularly issue to find the association rule from databasetransactions. After analysis and study the improved algorithm based onApriori on association rule mining, the paper According to thedisadvantage of Apriori algorithm present a optimization strategies areput forward to optimize the Apriori algorithm. At last, the experimentresults show that the speed of the mining is effectively improved in thisalgorithm.