摘要
在基于密度聚类算法的基础上,提出一种基于密度的快速异常检测方法 DBOD,该算法改变了DBSCAN算法对异常检测的被动处理方法,主动从异常出发,重点关注边界对象,实验证明该方法在检索速度方面具有明显优势.
This paper presents a fast outliers detection approach-DBOD （density-based outliers detection）, which is constructed on density-based clustering algorithms. Taking the place of DBSCAN algorithm＇s passive processing in outlier detection, this approach actively sets off from the outliers, and pays a big attention on boundary objects. Experiments were performed and the results indicate that this approach gains an advantage over the others in terms of the search speed.
出处
《伊犁师范学院学报（自然科学版）》
2012年第4期47-49,共3页
Journal of Yili Normal University：Natural Science Edition
