Association Rules Mining Algorithm Based on Information Gain Ratio Attribute Reduction

Abstract

In actual association rule mining, data sets collected from enterprises or real life often have some problems, such as a large amount of data missing or data redundancy, which greatly increases the spatial complexity of mining association rules and makes mining efficiency inefficient. Not only that, some actual data set contain hundreds or even more attributes. Not only does it take too long to mine association rules, but there are too many association rules obtained, making it difficult for users to distinguish which is more valuable information in practical applications. It is difficult to apply these data to actual enterprises to get greater benefits. In response to these problems, this paper proposes an association rule algorithm based on the FP-Growth association rule algorithm of information gain ratio attribute reduction to extract more valuable information and improve the efficiency of association rule mining. Finally, through experiments and comparisons, it is verified that the algorithm proposed in this paper can effectively mine the association rule information of multi-attribute data sets.