Integer Bayesian Network Classifiers

Integer Bayesian Network Classifiers

Abstract

This paper introduces integer Bayesian network classifiers (BNCs), i.e. BNCs with discrete valued nodes where parameters are stored as integer numbers. These networks allow for efficient implementation in hardware while maintaining a (partial) probabilistic interpretation under scaling. An algorithm for the computation of margin maximizing integer parameters is presented and its efficiency is demonstrated. The resulting parameters have superior classification performance compared to parameters obtained by simple rounding of double-precision parameters, particularly for very low number of bits.

Grafik Top
Authors
  • Tschiatschek, Sebastian
  • Paul, Karin
  • Pernkopf, Franz
Grafik Top
Shortfacts
Category
Paper in Conference Proceedings or in Workshop Proceedings (Paper)
Event Title
European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD)
Divisions
Data Mining and Machine Learning
Event Location
Nancy, France
Event Type
Conference
Event Dates
15.-19.09.2014
Series Name
Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2014. Lecture Notes in Computer Science
ISSN/ISBN
978-3-662-44844-1
Page Range
pp. 209-224
Date
15 September 2014
Export
Grafik Top