Integrating Bayesian networks and Simpson's paradox in data mining

Freitas, AA, McGarry, K and Correa, ES ORCID: https://orcid.org/0000-0002-5122-4384 2007, 'Integrating Bayesian networks and Simpson's paradox in data mining' , in: Causality and Probability in the Sciences , Texts in Philosophy, 5 , College Publications, United Kingdom, pp. 43-62.

PDF - Submitted Version Restricted to Repository staff only Download (255kB) | Request a copy

Abstract

This paper proposes to integrate two very different kinds of methods for data mining, namely the construction of Bayesian networks from data and the detection of occurrences of Simpson’s paradox. The former aims at discovering potentially causal knowledge in the data, whilst the latter aims at detecting surprising patterns in he data. By integrating these two kinds of methods we can hopefully discover patterns which are more likely to be useful to the user, a challenging data mining goal which is under-explored in the literature. The proposed integration method involves two approaches. The first approach uses the detection of occurrences of Simpson’s paradox as a preprocessing for a more effective construction of Bayesian networks; whilst the second approach uses the construction of a Bayesian network from data as a preprocessing for the detection of occurrences of Simpson’s paradox.

Item Type:	Book Section
Editors:	Russo, F and Williamson, J
Schools:	Schools > School of Computing, Science and Engineering > Salford Innovation Research Centre
Publisher:	College Publications
Series Name:	Texts in Philosophy
ISBN:	1904987354
Related URLs:	Organisation
Depositing User:	Dr Elon Correa
Date Deposited:	10 Feb 2017 15:11
Last Modified:	16 Feb 2022 18:11
URI:	https://usir.salford.ac.uk/id/eprint/41382

Actions (login required)

Edit record (repository staff only)