APPLICATION OF IMPROVED HIERARCHICAL CLUSTERING METHOD TO CLASSIFICATION OF CURVES

摘要：

An automatic clustering method for large-scale curves based on Hierarchical Clustering Method is pro posed in the paper. The method introduces similarity precision of curve classification as its terminal condition, and automatically classifies curves into several classifications according to the threshold of similarity precision, which meets precision request. Compared with traditional methods for curves clustering analysis, this method can automatically determine the number of classification, rather than depend on analysts’subjective determi nation, analysis results for curves clustering can be obtained automatically by this method and meet an any pre cision requirements set previously, Furthermore, analysis efficiency is therefore improved greatly. Satisfactory effect has been obtained when applying this method to analyze industry sales income curves.

关键词： Curves clustering Large-scale curves Hierarchical clustering method

作者: Ming Ye Huiwen Wang Lanhui Wang

作者单位: School of Economics and Management, Beihang University, Beijing 100083, China

会议类型: 国际会议

会议名称: The Ninth International Conference on Industrial Management(第九届工业管理国际会议 ICIM2008)

会议地点: 日本大阪

会议语种:英文

页码: 325-330

在线出版日期: 2008-09-16（万方平台首次上网日期，不代表论文的发表时间）