Extracting Credible Dependencies for Averaged One-Dependence Estimator Analysis
Joint Authors
Chi, BaoRong
Wang, ShuangCheng
Wang, LiMin
Li, XiongFei
Source
Mathematical Problems in Engineering
Issue
Vol. 2014, Issue 2014 (31 Dec. 2014), pp.1-11, 11 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2014-06-17
Country of Publication
Egypt
No. of Pages
11
Main Subjects
Abstract EN
Of the numerous proposals to improve the accuracy of naive Bayes (NB) by weakening the conditional independence assumption, averaged one-dependence estimator (AODE) demonstrates remarkable zero-one loss performance.
However, indiscriminate superparent attributes will bring both considerable computational cost and negative effect on classification accuracy.
In this paper, to extract the most credible dependencies we present a new type of seminaive Bayesian operation, which selects superparent attributes by building maximum weighted spanning tree and removes highly correlated children attributes by functional dependency and canonical cover analysis.
Our extensive experimental comparison on UCI data sets shows that this operation efficiently identifies possible superparent attributes at training time and eliminates redundant children attributes at classification time.
American Psychological Association (APA)
Wang, LiMin& Wang, ShuangCheng& Li, XiongFei& Chi, BaoRong. 2014. Extracting Credible Dependencies for Averaged One-Dependence Estimator Analysis. Mathematical Problems in Engineering،Vol. 2014, no. 2014, pp.1-11.
https://search.emarefa.net/detail/BIM-474094
Modern Language Association (MLA)
Wang, LiMin…[et al.]. Extracting Credible Dependencies for Averaged One-Dependence Estimator Analysis. Mathematical Problems in Engineering No. 2014 (2014), pp.1-11.
https://search.emarefa.net/detail/BIM-474094
American Medical Association (AMA)
Wang, LiMin& Wang, ShuangCheng& Li, XiongFei& Chi, BaoRong. Extracting Credible Dependencies for Averaged One-Dependence Estimator Analysis. Mathematical Problems in Engineering. 2014. Vol. 2014, no. 2014, pp.1-11.
https://search.emarefa.net/detail/BIM-474094
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-474094