An Improved Bat Algorithm based on Novel Initialization Technique for Global Optimization Problem

Bangyal, Waqas Haider; Ahmad, Jamil; Tayyab, Hafiz; Pervaiz, Sobia

doi:10.14569/ijacsa.2018.090723

Cited by 16 publications

(7 citation statements)

References 19 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training and validation image sets were randomly shuffled in order to have shuffled batches to avoid images correlation. The number of training epochs is selected based on the variations of the accuracy and validation error [26,27]. This is achieved by monitoring the validation error of each of the considered networks as it can be associated with optimization problems, i.e., if it is high, overfitting may occur.…”

Section: Resultsmentioning

confidence: 99%

Radiologists versus Deep Convolutional Neural Networks: A Comparative Study for Diagnosing COVID-19

Helwan

Ma'aitah²,

Hamdan

et al. 2021

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

The reverse transcriptase polymerase chain reaction (RT-PCR) is still the routinely used test for the diagnosis of SARS-CoV-2 (COVID-19). However, according to several reports, RT-PCR showed a low sensitivity and multiple tests may be required to rule out false negative results. Recently, chest computed tomography (CT) has been an efficient tool to diagnose COVID-19 as it is directly affecting the lungs. In this paper, we investigate the application of pre-trained models in diagnosing patients who are positive for COVID-19 and differentiating it from normal patients, who tested negative for coronavirus. The study aims to compare the generalization capabilities of deep learning models with two thoracic radiologists in diagnosing COVID-19 chest CT images. A dataset of 3000 images was obtained from the Near East Hospital, Cyprus, and used to train and to test the three employed pre-trained models. In a test set of 250 images used to evaluate the deep neural networks and the radiologists, it was found that deep networks (ResNet-18, ResNet-50, and DenseNet-201) can outperform the radiologists in terms of higher accuracy (97.8%), sensitivity (98.1%), specificity (97.3%), precision (98.4%), and F1-score (198.25%), in classifying COVID-19 images.

show abstract

Section: Resultsmentioning

confidence: 99%

Radiologists versus Deep Convolutional Neural Networks: A Comparative Study for Diagnosing COVID-19

Helwan

Ma'aitah²,

Hamdan

et al. 2021

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…e wavelength λ of the ultrasonic sound bursts in the range from 2 mm to 14 mm. Recently, an improved version of BA with novel initialization techniques was introduced in [63]. e Bats harness three rules: (i) to sense distance, background barriers, and dissimilarity between prey/food, bats use echolocation; (ii) bats fly randomly with a certain velocity to a specific position with a fixed frequency with varying wavelengths to search for prey; and (iii) the loudness changes from a large to minimum constant value [64].…”

Section: Modified Bat Algorithm For Global Optimizationmentioning

confidence: 99%

A Modified Bat Algorithm for Solving Large-Scale Bound Constrained Global Optimization Problems

Khan

Mehmood

Bakar

et al. 2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

In the last two decades, the field of global optimization has become very active, and, in this regard, many deterministic and stochastic algorithms were developed for solving various optimization problems. Among them, swarm intelligence (SI) is a stochastic algorithm that is more flexible and robust and has had the ability to find an optimum solution for high-dimensional optimization and search problems. SI-based algorithms are mainly inspired by the social behavior of fish schooling or bird flocking. Among the SI-based algorithms, Bat algorithm (BA) is one of the recently developed evolutionary algorithms. It employs an echolocation behavior of microbats by varying pulse rates of emission and loudness to perform their search process. In this paper, a modified Bat algorithm (MBA) is developed. The main focus of the MBA is to further enhance the exploration and exploitation search abilities of the original Bat algorithm. The performance of the modified Bat algorithm (MBA) is examined over the benchmark functions designed for evolutionary algorithms competition in the special session of 2005 IEEE Congress on Evolutionary Computation. The used benchmark functions include the unimodal, multimodal, and hybrid benchmark functions with high dimensionality. Furthermore, the impact analysis with respect to different values of temperatures is conducted by executing the proposed algorithm twenty-five times independently by using each benchmark function with different random seeds.

show abstract

“…An improved version of BA called I-BAT is proposed [61] to enhance the exploitation capabilities of BA. The authors used Torus distribution to initialize the bats in I-BAT and the controlled factor equal 0.1 is carried out for managing the search manner into the local area.…”

Section: Mutation Strategiesmentioning

confidence: 99%

An Overview of Mutation Strategies in Bat Algorithm

Bangyal¹,

Ahmad²,

Tayyab³

et al. 2018

ijacsa

Self Cite

View full text Add to dashboard Cite

Abstract-Bat algorithm (BA) is a population based stochastic search technique encouraged from the intrinsic manner of bee swarm seeking for their food source. BA has been mostly used to resolve diverse kind of optimization problems and one of major issue faced by BA is frequently captured in local optima meanwhile handling the complex real world problems. Many authors improved the standard BA with different mutation strategies but an exhausted comprehensive overview about mutation strategies is still lacking. This paper aims to furnish a concise and comprehensive study of problems and challenges that prevent the performance of BA. It has been tried to provide guidelines for the researchers who are active in the area of BA and its mutation strategies. The objective of this study is divided in two sections: primarily to display the improvement of BA with mutation strategies that may enhance the performance of standard BA up to great extent and secondly, to motivate the researchers and developers for using BA to solve the complex real world problems. This study presents a comprehensive survey of the various BA algorithms based on mutation strategies. It is anticipated that this survey would be helpful to study the BA algorithm in detail for the researcher.

show abstract

An Improved Bat Algorithm based on Novel Initialization Technique for Global Optimization Problem

Cited by 16 publications

References 19 publications

Radiologists versus Deep Convolutional Neural Networks: A Comparative Study for Diagnosing COVID-19

Radiologists versus Deep Convolutional Neural Networks: A Comparative Study for Diagnosing COVID-19

A Modified Bat Algorithm for Solving Large-Scale Bound Constrained Global Optimization Problems

An Overview of Mutation Strategies in Bat Algorithm

Contact Info

Product

Resources

About