Outlier effects on databases


Kaya A.

ADVANCES IN INFORMATION SYSTEMS, PROCEEDINGS, vol.3261, pp.88-95, 2004 (Journal Indexed in SCI) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 3261
  • Publication Date: 2004
  • Title of Journal : ADVANCES IN INFORMATION SYSTEMS, PROCEEDINGS
  • Page Numbers: pp.88-95

Abstract

Real data and databases always contain some kind of heterogenity or contamination, which is called "outliers". Outliers are defined as the few observations or records which appear to be inconsistent with the remainder group of the sample and more effective on prediction values. Isolated outliers may also have positive impact on the results of data analysis and data mining. In this study, we are concerned with outliers in time series which have two special cases, innovational outlier (10) and additive outlier (AO). The occurence of AO indicates that action is required, possibly to adjust the measuring instrument or at least to print an error message on the database. However, if 10 occurs, no adjustment of the measurement operation is required. At the end of the study, the results of the simulation and variance analysis on the produced data sets are emphasized.