点击次数：

论文类型：期刊论文

发表时间：2012-05-15

发表刊物：Journal of Computational Information Systems

收录刊物：EI、Scopus

卷号：8

期号：10

页面范围：4339-4346

ISSN号：15539105

摘要：Currently, patent data have gained increasing attention in the data mining area. However, the traditional method was set by many factors in patent classification, such as professional terms and high dimensions. In this article, we propose a word sense disambiguation model, which would effectively reduce the structure of the traditional machine word. The experimental result shows that the semantic disambiguation and feature reduction strategy can effectively improve the classification accuracy. ? 2011 by Binary Information Press.