| Abstract–Frequent pattern mining is a fundamental problem in data mining and knowledge discovery. The discovered patterns can be used as the input for analyzing association rules, mining sequential patterns, recognizing clusters, and so on. However, discovering frequent patterns in large scale datasets is an extremely time consuming task. Most research in the area of association rule discovery has focused on the subproblem of efficient frequent item set discovery (for example, Han, Pei, & Yin, 2000; Park, Chen, & Yu, 1995; Pei, Han, Lu, Nishio, Tang, & Yang, 2001; Savasere, Omiecinski, & Navathe, 1995). When seeking all associations that satisfy constraints on support and confidence, once frequent item sets have been identified, generating the association rules is trivial. Various efficient algorithms have been proposed and published on this problem in the last decade. However, there are a lot of challenges. In this paper, we research into a mathematical space coherent with full of methodologies, concepts, definitions, theories, theorems, algorithm, proofs, ... in order to apply to solving frequent patterns mining problem. The goal of the research is that the new well-researched algorithm have to be not theoretic, be easy and simple to understand, have low complexity, and resolve existing challenges in the world. Besides, the algorithm have to be easy to design and install, and apply with best effect. |
