Positive And Negative Association Rules Mining Algorithm In The Relational Data Mining

With the growth of database technology, popularity of network technology and updating of computer hardware, the capability of collecting data was improved rapidly. Hence, the capacity of storing data was enlarged hugely all over the world. To improve the understanding of vast data, data mining technology has improved rapidly.Association rules mining is one of important matters of data mining, which is advanced by Agrawal and the other in 1993. The purpose is analyzing the relation of items in transaction database. Later, because investigator improved and extended the prototype of question. At present, association rules technology has been applied to business, telecommunication, finance, agriculture, medical treatment and so on. It has brought a good effect. Relational database is an important form to store lots of information of production, management and scientific research. The increase of data quantum is very fast. Studying the efficient technology of mining association rules has a wide development in future.In this article, the classic Apriori algorithm for mining association rules is described in detail, the algorithm further thought is expressed through examples, and then for defects of Apriori algorithm, several improvements in the corresponding technology is briefly introduced. The FP-growth algorithm based on frequent pattern tree is also described in detail. The frequent pattern tree structure and the frequent patterns mining based on the FP-tree are specifically analysed.Combined with concrete examples the FP-growth algorithm idea is also analysed.Multi-level support is used in Algorithm. Positive association rules are generated from the frequent itemsets by the use of correlation, negative association rules are generated from the frequent itemsets and the non-frequent itemsets, and less valued rules are pruned by the use of correlation valueα. The effective rules are comparatively obtained by setting a reasonable mininterest and calculating the correlation and confidence level. The four multi-level confidence and chi-square test to test the correlation and independence of association rules are put forward.Association rule mining prototype instrument design, the main functions and implementation methods are proposed in the relational database. The main feature of the tool is that the data mining language is used as the core of design. On the one hand, high efficiency is ensured in a mature relational database theory and technology support ; on the other hand, mining tools and relational database system are seamlessly linked, so it becames more practical and convenient.