Search

Current clustering methods always have such problems: 1) High I/O cost and expensive maintenance; 2) Pre-specifying the uncertain parameter k; 3) Lacking good efficiency in treating arbitrary shape under very large data set environment. In this paper, we first present a hybrid-clustering algorithm to solve these problems. It combines both distance and density strategies, and makes full use of statistics information while keeping good cluster quality. The experimental results show that our algorithm outperforms other popular algorithms in terms of efficiency, cost, and even get much more speedup as the data size scales up.

XML is a standard that is widely applied in data representation and data exchange. However, as an important part of XML, DTD is not taken full advantage of in current applications. In this paper, a new method for clustering DTDs is presented, so that it can be used in XML document clustering. The two-level method clusters the elements in DTDs and DTDs separately. Element clustering forms the first level, and provides the element clusters, which is the generalization of relevant elements. DTD clustering utilizes the generalized information and forms the second level in the whole clustering process. The two-level method has the advantages that: 1) it takes into consideration both the content and the structure within the DTDs; 2) the generalized information about elements is more useful than the separated words in the vector model; 3) the two-level method facilitates the searching of outliers. The experiments show that this method is able to categorize the relevant DTDs effectively.