Improving Naive Bayes classifier using conditional probabilities
Version 2 2024-06-05, 03:23Version 2 2024-06-05, 03:23
Version 1 2019-07-09, 15:15Version 1 2019-07-09, 15:15
conference contribution
posted on 2024-06-05, 03:23 authored by S Taheri, Musa MammadovMusa Mammadov, AM BagirovNaive Bayes classifier is the simplest among Bayesian Network classifiers. It has shown to be very efficient on a variety of data classification problems. However, the strong assumption that all features are conditionally independent given the class is often violated on many real world applications. Therefore, improvement of the Naive Bayes classifier by alleviating the feature independence assumption has attracted much attention. In this paper, we develop a new version of the Naive Bayes classifier without assuming independence of features. The proposed algorithm approximates the interactions between features by using conditional probabilities. We present results of numerical experiments on several real world data sets, where continuous features are discretized by applying two different methods. These results demonstrate that the proposed algorithm significantly improve the performance of the Naive Bayes classifier, yet at the same time maintains its robustness. © 2011, Australian Computer Society, Inc.
History
Volume
121Pagination
63-68Location
Ballarat, Vic.Open access
- Yes
Start date
2011-12-01End date
2011-12-02ISSN
1445-1336ISBN-13
9781921770029Language
engPublication classification
EN.1 Other conference paperCopyright notice
2011, Australian Computer SocietyTitle of proceedings
AusDM11 : Proceedings of the 9th Australasian Data Mining Conference.Event
Australasian Data Mining. Conference (9th : 2011 : Ballarat, Victoria)Publisher
Australian Computer SocietyPlace of publication
Sydney, N.S.W.Series
Conferences in Research and Practice in Information Technology SeriesPublication URL
Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC