A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes

Naive Bayes (NB) is easy to construct but surprisingly effective, and it is one of the top ten classification algorithms in data mining. The conditional independence assumption of NB ignores the dependency between attributes, so its probability estimates are often suboptimal. Hidden naive Bayes (HNB...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Liangjun Yu, Shengfeng Gan, Yu Chen, Dechun Luo
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	Bayesian network hidden naive Bayes instance weighting Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/496adcfb2e3f4275bc51967387bbb127
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:496adcfb2e3f4275bc51967387bbb127
record_format	dspace
spelling	oai:doaj.org-article:496adcfb2e3f4275bc51967387bbb1272021-11-25T18:17:47ZA Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes10.3390/math92229822227-7390https://doaj.org/article/496adcfb2e3f4275bc51967387bbb1272021-11-01T00:00:00Zhttps://www.mdpi.com/2227-7390/9/22/2982https://doaj.org/toc/2227-7390Naive Bayes (NB) is easy to construct but surprisingly effective, and it is one of the top ten classification algorithms in data mining. The conditional independence assumption of NB ignores the dependency between attributes, so its probability estimates are often suboptimal. Hidden naive Bayes (HNB) adds a hidden parent to each attribute, which can reflect dependencies from all the other attributes. Compared with other Bayesian network algorithms, it offers significant improvements in classification performance and avoids structure learning. However, the assumption that HNB regards each instance equivalent in terms of probability estimation is not always true in real-world applications. In order to reflect different influences of different instances in HNB, the HNB model is modified into the improved HNB model. The novel hybrid approach called instance weighted hidden naive Bayes (IWHNB) is proposed in this paper. IWHNB combines instance weighting with the improved HNB model into one uniform framework. Instance weights are incorporated into the improved HNB model to calculate probability estimates in IWHNB. Extensive experimental results show that IWHNB obtains significant improvements in classification performance compared with NB, HNB and other state-of-the-art competitors. Meanwhile, IWHNB maintains the low time complexity that characterizes HNB.Liangjun YuShengfeng GanYu ChenDechun LuoMDPI AGarticleBayesian networkhidden naive Bayesinstance weightingMathematicsQA1-939ENMathematics, Vol 9, Iss 2982, p 2982 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Bayesian network hidden naive Bayes instance weighting Mathematics QA1-939
spellingShingle	Bayesian network hidden naive Bayes instance weighting Mathematics QA1-939 Liangjun Yu Shengfeng Gan Yu Chen Dechun Luo A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
description	Naive Bayes (NB) is easy to construct but surprisingly effective, and it is one of the top ten classification algorithms in data mining. The conditional independence assumption of NB ignores the dependency between attributes, so its probability estimates are often suboptimal. Hidden naive Bayes (HNB) adds a hidden parent to each attribute, which can reflect dependencies from all the other attributes. Compared with other Bayesian network algorithms, it offers significant improvements in classification performance and avoids structure learning. However, the assumption that HNB regards each instance equivalent in terms of probability estimation is not always true in real-world applications. In order to reflect different influences of different instances in HNB, the HNB model is modified into the improved HNB model. The novel hybrid approach called instance weighted hidden naive Bayes (IWHNB) is proposed in this paper. IWHNB combines instance weighting with the improved HNB model into one uniform framework. Instance weights are incorporated into the improved HNB model to calculate probability estimates in IWHNB. Extensive experimental results show that IWHNB obtains significant improvements in classification performance compared with NB, HNB and other state-of-the-art competitors. Meanwhile, IWHNB maintains the low time complexity that characterizes HNB.
format	article
author	Liangjun Yu Shengfeng Gan Yu Chen Dechun Luo
author_facet	Liangjun Yu Shengfeng Gan Yu Chen Dechun Luo
author_sort	Liangjun Yu
title	A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
title_short	A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
title_full	A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
title_fullStr	A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
title_full_unstemmed	A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes
title_sort	novel hybrid approach: instance weighted hidden naive bayes
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/496adcfb2e3f4275bc51967387bbb127
work_keys_str_mv	AT liangjunyu anovelhybridapproachinstanceweightedhiddennaivebayes AT shengfenggan anovelhybridapproachinstanceweightedhiddennaivebayes AT yuchen anovelhybridapproachinstanceweightedhiddennaivebayes AT dechunluo anovelhybridapproachinstanceweightedhiddennaivebayes AT liangjunyu novelhybridapproachinstanceweightedhiddennaivebayes AT shengfenggan novelhybridapproachinstanceweightedhiddennaivebayes AT yuchen novelhybridapproachinstanceweightedhiddennaivebayes AT dechunluo novelhybridapproachinstanceweightedhiddennaivebayes
_version_	1718411376834117632

A Novel Hybrid Approach: Instance Weighted Hidden Naive Bayes

Ejemplares similares