Arabian Journal for Science and Engineering (2021) 46:9237–9251
https://doi.org/10.1007/s13369-021-05674-9
RESEARCH ARTICLE-COMPUTER ENGINEERING AND COMPUTER SCIENCE
Hybrid Contractive Auto-encoder with Restricted Boltzmann Machine
For Multiclass Classification
Muhammad Aamir
1,2
· Nazri Mohd Nawi
1,3
· Fazli Wahid
4
· Muhammad Sadiq Hasan Zada
2
· M. Z. Rehman
1
·
Muhammad Zulqarnain
1
Received: 19 April 2020 / Accepted: 13 April 2021 / Published online: 23 June 2021
© King Fahd University of Petroleum & Minerals 2021
Abstract
Contractive auto-encoder (CAE) is a type of auto-encoders and a deep learning algorithm that is based on multilayer training
approach. It is considered as one of the most powerful, efficient and robust classification techniques, more specifically
feature reduction. The problem independence, easy implementation and intelligence of solving sophisticated problems make
it distinct from other deep learning approaches. However, CAE fails in data dimensionality reduction that cause difficulty to
capture the useful information within the features space. In order to resolve the issues of CAE, restricted Boltzmann machine
(RBM) layers have been integrated with CAE to enhance the dimensionality reduction and a randomized factor for hidden
layer parameters. The proposed model has been evaluated on four benchmark variant datasets of MNIST. The results have
been compared with four well-known multiclass class classification approaches including standard CAE, RBM, AlexNet and
artificial neural network. A considerable amount of improvement has been observed in the performance of proposed model as
compared to other classification techniques. The proposed CAE–RBM showed an improvement of 2–4% on MNIST(basic),
9–12% for MNIST(rot), 7–12% for MNIST(bg-rand) and 7–10% for MNIST(bg-img) dataset in term of final accuracy.
Keywords Contractive auto-encoder · Restricted Boltzmann machine · Classification · Mnist variants
B Muhammad Aamir
aamir@uthm.edu.my; m.aamir@derby.ac.uk
B Fazli Wahid
fazli.wahid@uoh.edu.pk
Nazri Mohd Nawi
nazri@uthm.edu.my
Muhammad Sadiq Hasan Zada
m.hassanzada@derby.ac.uk
M. Z. Rehman
syedzubair@uthm.edu.my
Muhammad Zulqarnain
zulqarnainmalik321@gmail.com
1
Faculty of Computer Science and Information Technology,
Universiti Tun Hussein Onn, Parit Raja, Malaysia
2
School of Electronics, Computing and Mathematics,
University of Derby, Derby, UK
3
Soft Computing and Data Mining Center, Universiti Tun
Hussein Onn, Parit Raja, Malaysia
4
Department of Information Technology, The University of
Haripur, Haripur, Khyber Pakhtunkhwa, Pakistan
Abbreviations
AE Auto-encoder
RBM Restricted Boltzmann machine
CAE Contractive auto-encoder
CAE–RBM Hybrid contractive auto-encoder–restricted
Boltzmann machine
ANN Artificial neural network
SVM Support vector machine
kNN k-Nearest neighbor
CNN Convolution neural network
DL Deep learning
ML Machine learning
ROC Receiver operating characteristic
CM Confusion matrix
MNIST Modified National Institute of Standards
and Technology (database)
rot MNIST random rotation digits
bg-rand Random noise background digits
bg-img Random background digits
123