Application And Research Of The Association Rules Based On Data Mining

With the development of science and technology, the big data has attracted more and more attention. A single data may be of no value, but when more and more data is accumulated, it will cause the qualitative change. The technology of people collecting data is continuing to progress. The technology of data mining is to extract valuable information from the data, which involves the integration of many disciplines technology and covers a very wide range. The association rules in data mining is produced and developed in the era of technology vigorous development.Along with the development of technology, now it has been widely applied to other industries. The core problem of association rules is to find the frequent itemsets. This thesis mainly introduces the classical algorithm Apriori. This algorithm needs to scan the database many rounds, and will produce a large number of candidate sets. Firstly this thesis proposes an improved Apriori algorithm, which can improve the efficiency of the Apriori algorithm. The FP-growth algorithm for constructing FP-tree might make the FP-tree structure is very lush, resulting in a waste of storage space and computing speed reduction defects. So secondly this thesis gives an improved method based chain addresses and split transaction database, while using divide and conquer strategy to find frequent item sets in order to improve the efficiency of data mining.This thesis has made the comparative experiments. By analysis of the experimental results, the improved algorithm can improve the efficiency of the original algorithm. Finally, the project has taken the verification for the improved Apriori algorithm and improved FP-Growth algorithm applied in the shopping basket data analysis to find the relationship in commodity effectively.