Hierarchical Document Clustering Using Fuzzy Association Rule Mining
Chunling Chen, Frank S. C. Tseng, Tyne Liang · 2008
In this paper, we will present an effective Fuzzy Frequent Itemset-Based Hierarchical Clustering (F2IHC) approach, which uses fuzzy frequent itemsets discovered by fuzzy association rule mining to improve the clustering accuracy of FIHC (Frequent Itemset-Based Hierarchical Clustering) method. Our approach can alleviate the deficiencies of most of the traditional document clustering methods in dealing with the problems of high dimensionality, large data size, and meaningful cluster labels. We have conducted experiments to evaluate our approach on Reuters 21578 dataset. The experimental results show that our approach not only absolutely retains the merits of FIHC, but also improves the document clustering accuracy quality as compared with the FIHC method.