An Effective Clustering Algorithm for Transaction Databases Based on K-Mean

Clustering is an important technique in machine learning, which has been successfully applied in many applications such as text and webpage classifications, but less in transaction database classification. A large organization usually has many branches and accumulates a huge amount of data in their branch databases called multidatabases. At present, the best way of mining multidatabases is, first, to classify them into different classes. In this paper, we redefine related concepts of transaction database clustering, and then in connection to the traditional clustering method, we propose a strategy of clustering transaction databases based on the k-mean. To prove that our strategy is effective and efficient, we implement the proposed algorithms. The results showed that the method of clustering transaction databases based on the k-mean is better than present methods.