Research On Association Rule Mining Algorithm And Its Application

With the development of information technology, especially the advent of the era of big data, large amounts of data have been accumulated. How to dig up useful knowledge from the rich data to change the way of our work and life, has become a hot research topic in the present and the future. As an important branch of data mining, association rule mining can be used to analyze the correlation between different items. The results can guide many commercial activities to make sound business decisions. Association rules mining plays a key role in processing business data.This thesis focuses on the research of association rule mining algorithm and its application. It describes the basic theory of association rules, followed by a detailed description of the classical association rule mining algorithm: Apriori and FP-growth, then compares the advantages and disadvantages of the two data mining algorithms. It puts forward the FP- growth algorithm of grouping and compression based on FP-growth algorithm, which is called GCFP-growth(Grouping and Compressing on Frequent-Pattern Growth). GCFP-growth algorithm improves the FP-growth algorithm from two aspects. Firstly, the data source is made packet by one attribute according to the purpose of the research, so the grouped data has at least one same item which can be omitted in mining process and added into result directly. This approach can reduce the attributes to be analyzed and relieve the processing pressure of a big amount of data in data mining. Secondly, this thesis achieves the compression of FP-tree by changing the sequence of the adjacent nodes and ignoring the intermediate nodes. Such a compression method can effectively reduce the generation of new nodes, so as to achieve the purpose of reducing occupied time and space.In this thesis, the superiority of GCFP-growth algorithm is validated and a recruitment information mining system is developed to reflect the practicality of GCFP-growth algorithm objectively. The core function of this system is to apply the GCFP-growth algorithm on recruitment information data. It will find interesting association rules through digging recruitment information and put forward feasible suggestions for the job seekers.