Mining causal outliers using gaussian Bayesian networks

Sakshi Babbar, Sanjay Chawla

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Outliers are often identified as data points which are ''rare'', ''isolated'', or far away from their nearest neighbours. In this paper we demonstrate that meaningful outliers, i.e., outliers which perhaps encode important or new information are those which violate causal relationships. We first build a Bayesian network which encode causal relationships between attributes and then identify those points as outliers which violate these causal relationships. Experiments on several data sets confirm that the outliers identified in this fashion are in some sense ''genuine'' as they reveal new information about the underlying data generating process.

Original languageEnglish
Title of host publicationProceedings - 2012 IEEE 24th International Conference on Tools with Artificial Intelligence, ICTAI 2012
Pages97-104
Number of pages8
DOIs
StatePublished - 2012
Externally publishedYes
Event2012 IEEE 24th International Conference on Tools with Artificial Intelligence, ICTAI 2012 - Athens, Greece
Duration: Nov 7 2012Nov 9 2012

Publication series

NameProceedings - International Conference on Tools with Artificial Intelligence, ICTAI
Volume1
ISSN (Print)1082-3409

Conference

Conference2012 IEEE 24th International Conference on Tools with Artificial Intelligence, ICTAI 2012
Country/TerritoryGreece
CityAthens
Period11/7/1211/9/12

Keywords

  • Bayesian networks
  • Causality and Outliers

Fingerprint

Dive into the research topics of 'Mining causal outliers using gaussian Bayesian networks'. Together they form a unique fingerprint.

Cite this