Abstract: A New Method for Patent Clustering Using Support Vector Clustering

The number of clusters has been needed for diverse clustering algorithms. Many researches for determining the number of clusters were published. Most of them focused on numeric data. The determination of the number of clusters is equally important to document data clustering. But, it is difficult to determine the number of clusters in the document data because the documents have diverse data types such as text and number. Also, the selection of the number by an evaluation measure is limited in document clustering. In this paper, we propose an ensemble method for determining the number of clusters in document clustering. This research will develop a new method for evaluating the determined number of clusters using voting approach of an ensemble method.