Dual instance and attribute weighting for Naive Bayes classification

Jia Wu; Shirui Pan; Zhihua Cai; Xingquan Zhu; Chengqi Zhang

doi:10.1109/IJCNN.2014.6889572

Dual instance and attribute weighting for Naive Bayes classification

Jia Wu, Shirui Pan, Zhihua Cai, Xingquan Zhu, Chengqi Zhang

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

10 Citations (Scopus)

Abstract

Naive Bayes (NB) network is a popular classification technique for data mining and machine learning. Many methods exist to improve the performance of NB by overcoming its primary weakness - the assumption that attributes are conditionally independent given the class, using techniques such as backwards sequential elimination and lazy elimination. Some weighting technologies, including attribute weighting and instance weighting, have also been proposed to improve the accuracy of NB. In this paper, we propose a dual weighted model, namely DWNB, for NB classification. In DWNB, we firstly employ an instance similarity based method to weight each training instance. After that, we build an attribute weighted model based on the new training data, where the calculation of the probability value is based on the embedded instance weights. The dual instance and attribute weighting allows DWNB to tackle the conditional independence assumption for accurate classification. Experiments and comparisons on 36 benchmark data sets demonstrate that DWNB outperforms existing weighted NB algorithms.

Original language	English
Title of host publication	2014 International Joint Conference on Neural Networks (IJCNN)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1675-1679
Number of pages	5
ISBN (Electronic)	9781479914845
DOIs	https://doi.org/10.1109/IJCNN.2014.6889572
Publication status	Published - 1 Jul 2014
Externally published	Yes
Event	2014 International Joint Conference on Neural Networks, IJCNN 2014 - Beijing, China Duration: 6 Jul 2014 → 11 Jul 2014

Conference

Conference	2014 International Joint Conference on Neural Networks, IJCNN 2014
Country/Territory	China
City	Beijing
Period	6/07/14 → 11/07/14

Keywords

Bayes methods
data mining
directed graphs
learning (artificial intelligence)
pattern classification
naive Bayes classification
machine learning
dual weighted model
DWNB
NB classification
instance similarity based method
training instance
attribute weighted model
training data
probability value
embedded instance weights
Niobium
Training
Accuracy
Training data
Data mining

Access to Document

10.1109/IJCNN.2014.6889572

Cite this

@inproceedings{3e3a4e871c5b4c2c9af2de9618968720,

title = "Dual instance and attribute weighting for Naive Bayes classification",

abstract = "Naive Bayes (NB) network is a popular classification technique for data mining and machine learning. Many methods exist to improve the performance of NB by overcoming its primary weakness - the assumption that attributes are conditionally independent given the class, using techniques such as backwards sequential elimination and lazy elimination. Some weighting technologies, including attribute weighting and instance weighting, have also been proposed to improve the accuracy of NB. In this paper, we propose a dual weighted model, namely DWNB, for NB classification. In DWNB, we firstly employ an instance similarity based method to weight each training instance. After that, we build an attribute weighted model based on the new training data, where the calculation of the probability value is based on the embedded instance weights. The dual instance and attribute weighting allows DWNB to tackle the conditional independence assumption for accurate classification. Experiments and comparisons on 36 benchmark data sets demonstrate that DWNB outperforms existing weighted NB algorithms.",

keywords = "Bayes methods, data mining, directed graphs, learning (artificial intelligence), pattern classification, naive Bayes classification, machine learning, dual weighted model, DWNB, NB classification, instance similarity based method, training instance, attribute weighted model, training data, probability value, embedded instance weights, Niobium, Training, Accuracy, Training data, Data mining",

author = "Jia Wu and Shirui Pan and Zhihua Cai and Xingquan Zhu and Chengqi Zhang",

year = "2014",

month = jul,

day = "1",

doi = "10.1109/IJCNN.2014.6889572",

language = "English",

pages = "1675--1679",

booktitle = "2014 International Joint Conference on Neural Networks (IJCNN)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "2014 International Joint Conference on Neural Networks, IJCNN 2014 ; Conference date: 06-07-2014 Through 11-07-2014",

}

Wu, J, Pan, S, Cai, Z, Zhu, X & Zhang, C 2014, Dual instance and attribute weighting for Naive Bayes classification. in 2014 International Joint Conference on Neural Networks (IJCNN). Institute of Electrical and Electronics Engineers (IEEE), pp. 1675-1679, 2014 International Joint Conference on Neural Networks, IJCNN 2014, Beijing, China, 6/07/14. https://doi.org/10.1109/IJCNN.2014.6889572

Dual instance and attribute weighting for Naive Bayes classification. / Wu, Jia; Pan, Shirui; Cai, Zhihua et al.

2014 International Joint Conference on Neural Networks (IJCNN). Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 1675-1679.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Dual instance and attribute weighting for Naive Bayes classification

AU - Wu, Jia

AU - Pan, Shirui

AU - Cai, Zhihua

AU - Zhu, Xingquan

AU - Zhang, Chengqi

PY - 2014/7/1

Y1 - 2014/7/1

N2 - Naive Bayes (NB) network is a popular classification technique for data mining and machine learning. Many methods exist to improve the performance of NB by overcoming its primary weakness - the assumption that attributes are conditionally independent given the class, using techniques such as backwards sequential elimination and lazy elimination. Some weighting technologies, including attribute weighting and instance weighting, have also been proposed to improve the accuracy of NB. In this paper, we propose a dual weighted model, namely DWNB, for NB classification. In DWNB, we firstly employ an instance similarity based method to weight each training instance. After that, we build an attribute weighted model based on the new training data, where the calculation of the probability value is based on the embedded instance weights. The dual instance and attribute weighting allows DWNB to tackle the conditional independence assumption for accurate classification. Experiments and comparisons on 36 benchmark data sets demonstrate that DWNB outperforms existing weighted NB algorithms.

AB - Naive Bayes (NB) network is a popular classification technique for data mining and machine learning. Many methods exist to improve the performance of NB by overcoming its primary weakness - the assumption that attributes are conditionally independent given the class, using techniques such as backwards sequential elimination and lazy elimination. Some weighting technologies, including attribute weighting and instance weighting, have also been proposed to improve the accuracy of NB. In this paper, we propose a dual weighted model, namely DWNB, for NB classification. In DWNB, we firstly employ an instance similarity based method to weight each training instance. After that, we build an attribute weighted model based on the new training data, where the calculation of the probability value is based on the embedded instance weights. The dual instance and attribute weighting allows DWNB to tackle the conditional independence assumption for accurate classification. Experiments and comparisons on 36 benchmark data sets demonstrate that DWNB outperforms existing weighted NB algorithms.

KW - Bayes methods

KW - data mining

KW - directed graphs

KW - learning (artificial intelligence)

KW - pattern classification

KW - naive Bayes classification

KW - machine learning

KW - dual weighted model

KW - DWNB

KW - NB classification

KW - instance similarity based method

KW - training instance

KW - attribute weighted model

KW - training data

KW - probability value

KW - embedded instance weights

KW - Niobium

KW - Training

KW - Accuracy

KW - Training data

KW - Data mining

UR - http://www.scopus.com/inward/record.url?scp=84908481882&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2014.6889572

DO - 10.1109/IJCNN.2014.6889572

M3 - Conference proceeding contribution

SP - 1675

EP - 1679

BT - 2014 International Joint Conference on Neural Networks (IJCNN)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2014 International Joint Conference on Neural Networks, IJCNN 2014

Y2 - 6 July 2014 through 11 July 2014

ER -

Dual instance and attribute weighting for Naive Bayes classification

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this