Mining association rules uses fuzzy weighted FP-growth

In data mining, the association rules are used to search for the relations of items of the transactions database. Following the data collected and stored, it can find values through association rules, and assist manager to proceed marketing strategies and plan market framework. In this paper, we attempt to use fuzzy partition method and decide membership function of quantitative values of each transaction item. Also, from managers we can reflect the importance of items as linguistic terms, which are transformed as fuzzy sets of weights. Next, fuzzy weighted frequent pattern growth is used to complete the process of data mining. The method above is expected to improve Apriori algorithm for its better efficiency of the whole association rules. An example is given to clearly illustrate the proposed approach. Finally, the experiment results are made to show the performance of the proposed methods.