期刊名称:Studia Universitatis Babes-Bolyai : Series Informatica
印刷版ISSN:1224-869X
出版年度:2006
卷号:LI
期号:02
页码:53-53
出版社:Babes-Bolyai University, Cluj-Napoca
摘要:The aim of this paper is to make a study on the in¡ãuence of
distance metrics for partitioning based aspect mining. For this purpose, we
comparatively present, from the aspect mining point of view, the results of
three algorithms in Aspect Mining, kAM ([3]), HAM ([6]) and GAAM ([5]),
for di.erent distance metrics. The evaluation is based on a set of quality
measure that we have previously de¡¥ned in [1] and [2], and a case study is
also reported. We introduce three di.erent criteria on which our study is
based.