Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Learning discrete Bayesian network parameters from continuous data streams: What is the best strategy?

Ratnapinda, P and Druzdzel, MJ (2015) Learning discrete Bayesian network parameters from continuous data streams: What is the best strategy? Journal of Applied Logic, 13 (4). 628 - 642. ISSN 1570-8683

[img] Plain Text (licence)
Available under License : See the attached license file.

Download (1kB)

Abstract

©2015 Elsevier B.V. All rights reserved. We compare three approaches to learning numerical parameters of discrete Bayesian networks from continuous data streams: (1) the EM algorithm applied to all data, (2) the EM algorithm applied to data increments, and (3) the online EM algorithm. Our results show that learning from all data at each step, whenever feasible, leads to the highest parameter accuracy and model classification accuracy. When facing computational limitations, incremental learning approaches are a reasonable alternative. While the differences in speed between incremental algorithms are not large (online EM is slightly slower), for all but small data sets online EM tends to be more accurate than incremental EM.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: Article
Status: Published
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Ratnapinda, P
Druzdzel, MJdruzdzel@pitt.eduDRUZDZEL
Date: 1 January 2015
Date Type: Publication
Journal or Publication Title: Journal of Applied Logic
Volume: 13
Number: 4
Page Range: 628 - 642
DOI or Unique Handle: 10.1016/j.jal.2015.03.007
Schools and Programs: School of Information Sciences > Information Science
Refereed: Yes
ISSN: 1570-8683
Date Deposited: 02 Jul 2015 14:27
Last Modified: 30 Oct 2017 22:55
URI: http://d-scholarship.pitt.edu/id/eprint/25531

Metrics

Monthly Views for the past 3 years

Plum Analytics

Altmetric.com


Actions (login required)

View Item View Item