The common feature of transaction datasets is that it is very huge in size, so it is important to develop a technique for dataset reduction. The process of dataset reduction must not change the features of the original dataset; this will increase the effectiveness and efficiency of extracting association rules from these datasets without affecting the original data. Disjoint clusters that have different number of transactions will be introduced in order to minimize the search space, this in turn will decrease the time required to mine the desired rules by dealing with each cluster individually. The support and confidence measures will be used to determine the frequent item sets and exclude the others.

Association rules mining (ARM) algorithms have been extensively researched in the last decade. Therefore, numerous algorithms were proposed to discover frequent itemsets and then mine association rules. This paper will present an efficient ARM algorithm by proposing a new technique to generate association rules from a huge set of items, which depends on the concepts of clustering and graph data structure, this new algorithm will be named clustering and graph-based rule mining (CGAR). The CGAR method is to create a cluster table by scanning the database only once, and then clustering the transactions into clusters according to their length. The frequent 1-itemsets will be extracted directly by scanning the cluster table. To obtain frequent k-itemsets, where k ges 2, we build directed graphs for each cluster in the case of very huge amount of transactions. This approach reduces main memory requirement since it considers only a small cluster at a time and hence it is scalable for any large size of the database. Experiments show that our algorithm outperforms other rule mining algorithms.

Mining association rules is an essential task for knowledge discovery. Past transaction data can be analyzed to discover customer behaviors such that the quality of business decision can be improved. The approach of mining association rules focuses on discovering large itemsets, which are groups of items that appear together in an adequate number of transactions. In this paper, we propose a graph-based approach (DGARM) to generate Boolean association rules from a large database of customer transactions. This approach scans the database once to construct an association graph and then traverses the graph to generate all large itemsets. Practical evaluations show that the proposed algorithm outperforms other algorithms which need to make multiple passes over the database.

There has been an increased demand for structured data mining. Graphs are among the most extensively researched data structures in discrete mathematics and computer science. Thus, it should come as no surprise that graph-based data mining has gained popularity in recent years. Graph-based methods for a transaction database are necessary to transform all the information into a graph form to conveniently extract more valuable information to improve the decision-making process. Graph-based data mining can reveal and measure process insights in a detailed structural comparison strategy that is ready for further analysis without the loss of significant details. This paper analyzes the similarities and differences among four of the most popular graph-based methods that is applied to mine rules from transaction databases by abstracting them out as a concrete high-level interface and connecting them into a common space.