Empirical bayesian binary classification forests using bootstrap prior

In this paper, we present a new method called Empirical Bayesian Random Forest (EBRF) for binary classification problem. The prior ingredient for the method was obtained using the bootstrap prior technique. EBRF addresses explicitly low accuracy problem in Random Forest (RF) classifier when the numb...

Full description

Bibliographic Details
Main Authors: Olaniran, Oyebayo Ridwan, Abdullah, Mohd Asrul Affendi, Gopal Pillay, Khuneswari A/P, Olaniran, Saidat Fehintola
Format: Article
Published: Science Publishing Corporation 2018
Subjects:
Online Access:http://eprints.uthm.edu.my/3676/
_version_ 1848888085398421504
author Olaniran, Oyebayo Ridwan
Abdullah, Mohd Asrul Affendi
Gopal Pillay, Khuneswari A/P
Olaniran, Saidat Fehintola
author_facet Olaniran, Oyebayo Ridwan
Abdullah, Mohd Asrul Affendi
Gopal Pillay, Khuneswari A/P
Olaniran, Saidat Fehintola
author_sort Olaniran, Oyebayo Ridwan
building UTHM Institutional Repository
collection Online Access
description In this paper, we present a new method called Empirical Bayesian Random Forest (EBRF) for binary classification problem. The prior ingredient for the method was obtained using the bootstrap prior technique. EBRF addresses explicitly low accuracy problem in Random Forest (RF) classifier when the number of relevant input variables is relatively lower compared to the total number of input variables. The improvement was achieved by replacing the arbitrary subsample variable size with empirical Bayesian estimate. An illustration of the proposed, and existing methods was performed using five high-dimensional microarray datasets that emanated from colon, breast, lymphoma and Central Nervous System (CNS) cancer tumours. Results from the data analysis revealed that EBRF provides reasonably higher accuracy, sensitivity, specificity and Area Under Receiver Operating Characteristics Curve (AUC) than RF in most of the datasets used.
first_indexed 2025-11-15T20:04:40Z
format Article
id uthm-3676
institution Universiti Tun Hussein Onn Malaysia
institution_category Local University
last_indexed 2025-11-15T20:04:40Z
publishDate 2018
publisher Science Publishing Corporation
recordtype eprints
repository_type Digital Repository
spelling uthm-36762021-11-21T07:02:08Z http://eprints.uthm.edu.my/3676/ Empirical bayesian binary classification forests using bootstrap prior Olaniran, Oyebayo Ridwan Abdullah, Mohd Asrul Affendi Gopal Pillay, Khuneswari A/P Olaniran, Saidat Fehintola T57.6-57.97 Operations research. Systems analysis In this paper, we present a new method called Empirical Bayesian Random Forest (EBRF) for binary classification problem. The prior ingredient for the method was obtained using the bootstrap prior technique. EBRF addresses explicitly low accuracy problem in Random Forest (RF) classifier when the number of relevant input variables is relatively lower compared to the total number of input variables. The improvement was achieved by replacing the arbitrary subsample variable size with empirical Bayesian estimate. An illustration of the proposed, and existing methods was performed using five high-dimensional microarray datasets that emanated from colon, breast, lymphoma and Central Nervous System (CNS) cancer tumours. Results from the data analysis revealed that EBRF provides reasonably higher accuracy, sensitivity, specificity and Area Under Receiver Operating Characteristics Curve (AUC) than RF in most of the datasets used. Science Publishing Corporation 2018 Article PeerReviewed Olaniran, Oyebayo Ridwan and Abdullah, Mohd Asrul Affendi and Gopal Pillay, Khuneswari A/P and Olaniran, Saidat Fehintola (2018) Empirical bayesian binary classification forests using bootstrap prior. International Journal of Engineering and Technology, 7 (4.3). pp. 170-175. ISSN 2227-524X
spellingShingle T57.6-57.97 Operations research. Systems analysis
Olaniran, Oyebayo Ridwan
Abdullah, Mohd Asrul Affendi
Gopal Pillay, Khuneswari A/P
Olaniran, Saidat Fehintola
Empirical bayesian binary classification forests using bootstrap prior
title Empirical bayesian binary classification forests using bootstrap prior
title_full Empirical bayesian binary classification forests using bootstrap prior
title_fullStr Empirical bayesian binary classification forests using bootstrap prior
title_full_unstemmed Empirical bayesian binary classification forests using bootstrap prior
title_short Empirical bayesian binary classification forests using bootstrap prior
title_sort empirical bayesian binary classification forests using bootstrap prior
topic T57.6-57.97 Operations research. Systems analysis
url http://eprints.uthm.edu.my/3676/