作 者: ;
机构地区: 广东广播电视大学
出 处: 《广东广播电视大学学报》 2009年第3期109-112,共4页
摘 要: 事务识别在Web日志挖掘步骤中至关重要,如果没有高效的识别算法,挖掘的效率将大打折扣。本文详细研究了事务识别的两种模型,介绍并改进了基于最大向前引用模型的识别方法MFP算法,同时对这两种方法进行了比较,结果表时,改进后的MFP算法具有更高的挖掘效率。 Transaction identification plays a key role in the process of web log mining. Without efficient identification arithmetic, the efficiency of web log mining will be greatly reduced, This paper begins with a detailed analysis on transaction identification two models, and follow by the introduction of identification method based on the maximal forward path(MFP), and concluded with an exploration of examples .