首页>
外国专利>
METHOD OF CLUSTERING MULTI-DIMENSIONAL RELATED DATA IN A COMPUTER DATABASE BY COMBINING THE TWO VERTICES OF A GRAPH CONNECTED BY AN EDGE HAVING THE HIGHEST SCORE
METHOD OF CLUSTERING MULTI-DIMENSIONAL RELATED DATA IN A COMPUTER DATABASE BY COMBINING THE TWO VERTICES OF A GRAPH CONNECTED BY AN EDGE HAVING THE HIGHEST SCORE
展开▼
机译:通过结合具有最高得分的边缘连接的两个图形在计算机数据库中聚类多维相关数据的方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
A method of clustering multi-dimensional relateddata performed by identifying features from a collectionof data, each of said features being represented by avertex (31, 32, 33, 34, 35), selecting pairs of featuresthat it is desirable to cluster together, connecting the pairof vertices representing every selected pair of featuresby an edge (40, 41, 42, 43), assigning a score to everyedge according to a predetermined formula, selectingan edge having the highest score, creating a new vertexby merging the vertices connected by the selected edge(17), creating new edges between the new vertex andthe vertices previously connected to the merged vertices,and repeating this procedure until every edge has apredetermined score.
展开▼