An Efficient Hash-based Association Rule Mining Approach for Document Clustering

Document clustering is one of the important research issues in the field of text mining, where the documents are grouped without predefined categories or labels. High dimensionality is a major challenge in document clustering. Some of the recent algorithms address this problem by using frequent term sets for clustering. This paper proposes a new methodology for document clustering based on Association Rules Mining. Our approach consists of three phases: the text preprocessing phase, the association rule mining phase, and the document clustering phase. An efficient Hash-based Association Rule Mining in Text (HARMT) algorithm is used to overcome the drawbacks of Apriori algorithm. The generated association rules are used for obtaining the partition, and grouping the partition that have the same documents. Furthermore, the resultant clusters are effectively obtained by grouping the partition by means of derived keywords. Our approach can reduce the dimension of the text efficiently for very large text documents, thus it can improve the accuracy and speed of the clustering algorithm. Key-Words: Document Clustering, knowledge discovery, Hashing, Association Rule Mining, Text Documents, Text Mining.