Investigation of outlier detection algorithm

The proposed outlier factor was used to analyze the multidimensional data sets regarding outlier detection. The paper describes two kinds of investigation: the influence of omitting some part of distances between data points, and the influence of additional (noisy) points to outlier detection quali...

Full description

Saved in:
Bibliographic Details
Main Author: Vydūnas Šaltenis
Format: Article
Language:English
Published: Vilnius University Press 2005-12-01
Series:Lietuvos Matematikos Rinkinys
Subjects:
Online Access:https://www.journals.vu.lt/LMR/article/view/29330
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The proposed outlier factor was used to analyze the multidimensional data sets regarding outlier detection. The paper describes two kinds of investigation: the influence of omitting some part of distances between data points, and the influence of additional (noisy) points to outlier detection quality.  The results demonstrate the possibilities to improve the performance of computation and the stability of the outlier detection algorithm.
ISSN:0132-2818
2335-898X