Mine-first association rule mining: An integration of independent frequent patterns in distributed environments

Abstract

Association rule mining is a widely used data mining technique in various domains. It enables the identification of trends, frequent patterns, and relationships among the data. This study introduced a new method for mining association rules independently from multiple data sources. It combined the frequent patterns obtained from each data source to discover frequent patterns applicable across the distributed environment. The model can also be extended to generate the rules with the specified target. The proposed method’s performance is compared to that of the traditional association rule mining method. The experimental results demonstrate that while the generated rules may not be identical to those produced by the traditional method, the proposed model offers better transparency and memory utilization in association rule generation. In addition, the model uncovers meaningful relationships, allowing decision-makers to access the frequent patterns for the individual data sources and the entire data across the environment.