Skip to main content

Zahra Beheshti

Islamic Azad University, Najafabad Branch, Computer Engineering, Faculty Member

Universiti Teknologi Malaysia - UTM, Computer Science and Information System, Department Member

Followers

393

Following

45

Co-author

1

Public Views

Zahra Beheshti received her PhD and Post-Doc in Artificial Intelligence from Universiti Teknologi Malaysia (UTM), Malaysia. She is Assistant Professor at Islamic Azad University Najafabad Branch. Her current research interests include Soft Computing and its Applications, Pattern Recognition, and Design of Algorithms. She can be reached at [email protected] or [email protected].
Zahra Beheshti was ranked in Stanford University study as the world's top 2% of scientists in 2020.

less

School of the Art Institute of Chicago

Jordi Vallverdú

Universitat Autònoma de Barcelona

Atle Mikkola Kjøsen

Western University Canada

Kansas State University

Geisinger Health System

Near East University

Université Laval

University of Southampton

PALIMOTE JUSTICE

RIVERS STATE POLYTECHNIC

Wendy Lynne Lee

Bloomsburg University of Pennsylvania

Interests

Uploads

Papers by Zahra Beheshti

An Agglomerative Hierarchical Clustering Framework for Improving the Ensemble Clustering Process

Cybernetics and Systems

An Enhanced Forwarding Method based on Intelligent Water Drops Algorithm in Named Data Network

Majlesi Journal of Electrical Engineering, 2021

Electromagnetic field optimization: A physics-inspired metaheuristic optimization algorithm

Swarm and Evolutionary Computation, 2016

This paper presents a physics-inspired metaheuristic optimization algorithm, known as Electromagn... more This paper presents a physics-inspired metaheuristic optimization algorithm, known as Electromagnetic Field Optimization (EFO). The proposed algorithm is inspired by the behavior of electromagnets with different polarities and takes advantage of a nature-inspired ratio, known as the golden ratio. In EFO, a possible solution is an electromagnetic particle made of electromagnets, and the number of electromagnets is determined by the number of variables of the optimization problem. EFO is a populationbased algorithm in which the population is divided into three fields (positive, negative, and neutral); attraction-repulsion forces among electromagnets of these three fields lead particles toward global minima. The golden ratio determines the ratio between attraction and repulsion forces to help particles converge quickly and effectively. The experimental results on 30 high dimensional CEC 2014 benchmarks reflect the superiority of EFO in terms of accuracy and convergence speed over other state-of-the-art optimization algorithms.

Fuzzy sign-aware diffusion models for influence maximization in signed social networks

Information Sciences, Oct 1, 2023

FSCN: A Novel Forwarding Method based on Shannon Entropy and COPRAS Decision Process in Named Data Networking

One of the next-generation Information-Centric Networking (ICN) architectures for the future Inte... more One of the next-generation Information-Centric Networking (ICN) architectures for the future Internet is Named Data Networking (NDN). The appropriate output interface (forwarding) has recently emerged as a key challenge affecting the performance of this network and has received a lot of attention from researchers due to its importance. Although efforts have been made through the methods introduced so far to address this challenge, the neglect of effective network load balancing parameters, non-dynamic weighting, and the use of only a few criteria have been suggested for selecting the output interface have made it vital to further improve the network delay, congestion, and ultimately the network's throughput. Regarding this matter, a new strategy is presented in this article based on the COPRAS Decision-Making Process with the dynamic weighting of Shannon's entropy called FSCN for transmitting request packets through the optimal output interface. The essential parameters iden...

Binary Accelerated Particle Swarm Algorithm (BAPSA) for discrete optimization problems

Journal of Global Optimization, Dec 14, 2012

Revolutionizing Learning Technology Based on Information Technology

The information society is a utopia of the scholars' target. To achieve information society k... more The information society is a utopia of the scholars' target. To achieve information society knowledge production and distribution is necessary and the starting point is education. Information technology and communication can develop basic steps involved in public education and promote community knowledge. One of the discussions considered by the advocaters of the development of information technology in education is that they can raise educational opportunities for different classes of society, especially the poor. Experience shows that the latter does not happen consistently and if the government does not finance for the access of the poor to resources, information technology gap which called "digital gap" increases daily. Our country needs to have a new attitude towards the education of the young eager to go to schools and universities, the uneducated, and the laborers in order to increase their efficiency in specific specialties and to improve training programs in s...

UTF: Upgrade transfer function for binary meta-heuristic algorithms

Applied Soft Computing, 2021

In the real world, many optimization problems are discrete and very complex to solve. Some of the... more In the real world, many optimization problems are discrete and very complex to solve. Some of them are in the class of NP-hard problems and their search spaces grow exponentially with the problem size. As a result, an exhaustive search will be impractical using exact algorithms. In the last decades, meta-heuristic algorithms as approximate algorithms have shown superior performance in solving these problems. The majority of these algorithms have been designed for continuous search spaces and are not able to solve binary optimization problems. Therefore, a transfer function is applied to convert the continuous search space to the binary one. The performance of such binary algorithms depends on their ability of exploration, exploitation and transfer function. Several transfer functions have been introduced so far but they have shown poor exploration and exploitation in solving some problems. In this study, a novel adaptive transfer function, based on two linear functions, is proposed to overcome the shortcomings of existing transfer functions. The proposed method called upgrade transfer function (UTF) adapts itself during running the algorithm to switch from exploration to exploitation. This capability also covers disadvantages of meta-heuristic algorithms in terms of poor exploration and exploitation. The performance of UTF has been evaluated by three discrete optimization problems: function optimization, feature selection and the 0–1 multi-knapsack problem (MKP). The results of binary particle swarm optimization (BPSO), binary artificial bee colony (BABC) and several improved BPSO and BABC have been compared with those of UTF–BPSO and UTF–BABC using function optimization problems. Also, the efficiency of UTF–BPSO and UTF–BABC and some binary meta-heuristic algorithms such as binary salp swarm algorithm (BSSA) and binary gray wolf optimization (bGWO), binary dragon algorithm (BDA), binary multi-neighborhood artificial bee colony (BMNABC), binary hybrid topology particle swarm optimization quadratic interpolation (BHTPSO-QI), binary ant lion optimizer (bALO) and binary gravitational search algorithm (BGSA) have been evaluated by feature selection problems. Moreover, UTF and seventeen transfer functions have been applied in original PSO, ABC, SSA and GWO algorithms to solve low and high dimensions 0–1 MKP benchmark instances. The results showed that the new transfer function significantly enhances the performance of algorithms to achieve the best solution in the binary search space.

Software Engineering and Information System Applications in life

In recent century, every country based on its political, social and cultural situation, has a spe... more In recent century, every country based on its political, social and cultural situation, has a special goal to follow in order to develop city management, in turn, to improve the quality of services without any limit. On the top of these services lie life standards and the presence of citizen in decision making. The best way to actualize this goal is the existence of cities with e-management. In traditional cities, citizens consume too much on useless repetitive activities. Banking, official, business, educational, scientific, recreational, and political affairs as well as health care are activities that a person may do every day. Due to the fact that most cities provided with emanagement, the citizen's access possibility to different facilities in offices and places is provided round-the-clock. The aim of this paper is to investigate e-city properties and the applications of software engineering and information technology in an e-city. In this regard, basic challenges are invest...

Improved centripetal accelerated particle swarm optimization

Particle swarm optimization (PSO) is a bio-inspired optimization algorithm that imitates the soci... more Particle swarm optimization (PSO) is a bio-inspired optimization algorithm that imitates the social behavior of bird flocking, fish schooling and swarm theory. Although PSO has a simple concept and is easy to implement, it may converge prematurely because of its poor exploration when solving complex multimodal problems. Centripetal accelerated particle swarm optimization (CAPSO) is an enhanced particle swarm optimization (PSO) scheme combined with Newton’s laws of motion. It has shown an effective algorithm in solving optimization problems however; its performance can be enhanced similar to other evolutionary computation algorithms. In this study, an improved CAPSO (ICAPSO) and improved binary CAPSO (IBCAPSO) are introduced to accelerate the learning procedure and convergence rate of optimization problems in the real and binary search spaces. The proposed algorithms are evaluated by twenty high-dimensional complex benchmark functions. The results showed that the methods substantiall...

R-GWO: Representative-based grey wolf optimizer for solving engineering problems

Applied Soft Computing, 2021

Abstract The grey wolf optimizer (GWO) is a well-known nature-inspired algorithm, which shows a s... more Abstract The grey wolf optimizer (GWO) is a well-known nature-inspired algorithm, which shows a sufficient performance for solving various optimization problems. However, it suffers from low exploration and population diversity because its optimization process is only based on the best three wolves greedily, and the information of other wolves does not consider. In this paper, a representative-based grey wolf optimizer (R-GWO) is proposed to tackle with these weaknesses of the GWO. The R-GWO introduces a search strategy named representative-based hunting (RH) a combination of three effective trial vectors inspired by alpha wolves’ behaviors to improve the exploration and diversity of the population. The RH search strategy utilizes a representative archive to reduces the greediness and enhance the diversity of solutions, and it can also strike balance between the exploration and exploitation using a non-linear control parameter. The performance and applicability of the proposed R-GWO were evaluated on CEC 2018 benchmark functions and six engineering design problems. The results were compared by eight state-of-the-art metaheuristic algorithms: PSO, KH, GWO, WOA, EEGWO, BOA, HHO, and HGSO. Moreover, the results were statistically analyzed by three test Wilcoxon rank-sum, Friedman and mean absolute error (MAE). The performance results show that on all 29 functions with dimensions 30, 50, and 100, the R-GWO is superior to the competitor algorithms except on function 27 on all dimensions and function 22 on dimension 30. The proposed R-GWO is the most effective algorithm compared with competitor algorithms, with an overall effectiveness of 95.4%. The experimental and statistical results show that the R-GWO is competitive and superior to compared algorithms and can solve engineering design problems better than competitor algorithms.

A novel x-shaped binary particle swarm optimization

Soft Computing, 2020

Definitive optimization algorithms are not able to solve high-dimensional optimization problems b... more Definitive optimization algorithms are not able to solve high-dimensional optimization problems because the search space grows exponentially with the problem size and an exhaustive search will be impractical. Therefore, approximate algorithms are applied to solve them. A category of approximate algorithms are meta-heuristic algorithms. They have shown an acceptable efficiency to solve these problems. Among them, particle swarm optimization (PSO) is one of the well-known swarm intelligence algorithms to optimize continuous problems. A transfer function is applied in this algorithm to convert the continuous search space to the binary one. The role of the transfer function in binary PSO (BPSO) is very important to enhance its performance. Several transfer functions have been proposed for BPSO such as S-shaped, V-shaped, linear and other transfer functions. However, BPSO algorithm can sometimes find local optima or show slow convergence speed in some problems because of using the velocity of PSO and these transfer functions. In this study, a novel transfer function called x-shaped BPSO (XBPSO) is proposed to increase exploration and exploitation of BPSO in the binary search space. The transfer function uses two functions and improved rules to generate a new binary solution. The proposed method has been run on 33 benchmark instances of the 0–1 multidimensional knapsack problem (MKP), two discrete maximization functions and 23 minimization functions. The results have been compared with some well-known BPSO and discrete meta-heuristic algorithms. The results showed that x-shaped transfer function considerably increased the solution accuracy and convergence speed in BPSO algorithm. The average error of compared algorithms on all 0–1 MKP benchmark instances indicated that XBPSO has the minimum error of 8.9%. Also, the mean absolute error (MAE) obtained by XBPSO on two discrete maximization functions is 0.45. Moreover, the proposed transfer function provides superior solutions in 18 functions from 23 minimization functions.

A new rainfall forecasting model using the CAPSO algorithm and an artificial neural network

Neural Computing and Applications, 2015

Artificial neural networks (ANNs) are being used increasingly to forecast rainfall. In this study... more

Improvement of Multi-Layer Perceptron (MLP) training using optimization algorithms

Evolutionary Algorithms for Neural Network Learning Enhancement

Artificial Neural Network (ANN) is one of the modern computational methods proposed to solve the ... more Artificial Neural Network (ANN) is one of the modern computational methods proposed to solve the majority of real world problems. Back-Propagation (BP) algorithm (as a gradient descend method) is one of the most popular methods for ANN training. However, there are some unavoidable disadvantages such as slow convergence speed, easily gets into partial extreme value and infirm global searching capability. To solve these problems, many solutions have been presented so far. Among them, Evolutionary Algorithms (EAs) have shown a good performance in this regard. EAs use some mechanisms inspired by biological evolution to find an optimization solution. Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Imperialist Competitive Algorithm (ICA) are in the class of algorithms. In this study, the mentioned optimization algorithms are chosen and applied in feed-forward neural network to enhance the learning process in terms of convergence rate and classification accuracy.

A review of population-based meta-heuristic algorithms

Enhancement of artificial neural network learning using centripetal accelerated particle swarm optimization for medical diseases diagnosis

Soft Computing, 2013

In recent decades, artificial neural networks (ANNs) have been extensively applied in different a... more In recent decades, artificial neural networks (ANNs) have been extensively applied in different areas such as engineering, medicine, business, education, manufacturing and so on. Nowadays, ANNs are as a hot research in medicine especially in the fields of medical disease diagnosis. To have a high efficiency in ANN, selection of an appropriate architecture and learning algorithm is very important. ANN learning is a complex task and an efficient learning algorithm has a significant role to enhance ANN performance. In this paper, a new meta-heuristic algorithm, centripetal accelerated particle swarm optimization (CAPSO), is applied to evolve the ANN learning and accuracy. The algorithm is based on an improved scheme of particle swarm algorithm and Newton's laws of motion. The hybrid learning of CAPSO and multi-layer perceptron (MLP) network, CAPSO-MLP, is used to classify the data of nine standard medical datasets of Hepatitis, Heart Disease, Pima Indian Diabetes, Wisconsin Prognostic Breast Cancer, Parkinson's disease, Echocardiogram, Liver Disorders, Laryngeal 1 and Acute Inflammations. The performance of CAPSO-MLP is compared with those of PSO, gravitational search algorithm and imperialist Communicated by D. Liu.

Binary Accelerated Particle Swarm Algorithm (BAPSA) for discrete optimization problems

Journal of Global Optimization, 2012

Your article is protected by copyright and all rights are held exclusively by Springer Science +B... more Your article is protected by copyright and all rights are held exclusively by Springer Science +Business Media New York. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com".

CAPSO: Centripetal accelerated particle swarm optimization

Information Sciences, 2014

ABSTRACT Meta-heuristic search algorithms are developed to solve optimization problems. Such algo... more ABSTRACT Meta-heuristic search algorithms are developed to solve optimization problems. Such algorithms are appropriate for global searches because of their global exploration and local exploitation abilities. Swarm intelligence (SI) algorithms comprise a branch of meta-heuristic algorithms that imitate the behavior of insects, birds, fishes, and other natural phenomena to find solutions for complex optimization problems. In this study, an improved particle swarm optimization (PSO) scheme combined with Newton’s laws of motion, the centripetal accelerated particle swarm optimization (CAPSO) scheme, is introduced. CAPSO accelerates the learning and convergence of optimization problems. In addition, the binary mode of the proposed algorithm, binary centripetal accelerated particle swarm optimization (BCAPSO), is introduced for binary search spaces. These algorithms are evaluated using nonlinear benchmark functions, and the results are compared with the gravitational search algorithm (GSA) and PSO in both the real and the binary search spaces. Moreover, the performance of CAPSO in solving the functions is compared with some well-known PSO algorithms in the literature. The experimental results showed that the proposed methods enhance the performance of PSO in terms of convergence speed, solution accuracy and global optimality.

A Review of Emotional Learning And It's Utilization in Control Engineering

Int. J. Advance. Soft Comput. Appl, 2010

In mammalians, brain Limbic System is responsible for emotional process. An important part of mot... more

An Agglomerative Hierarchical Clustering Framework for Improving the Ensemble Clustering Process

Cybernetics and Systems

An Enhanced Forwarding Method based on Intelligent Water Drops Algorithm in Named Data Network

Majlesi Journal of Electrical Engineering, 2021

Electromagnetic field optimization: A physics-inspired metaheuristic optimization algorithm

Swarm and Evolutionary Computation, 2016

This paper presents a physics-inspired metaheuristic optimization algorithm, known as Electromagn... more This paper presents a physics-inspired metaheuristic optimization algorithm, known as Electromagnetic Field Optimization (EFO). The proposed algorithm is inspired by the behavior of electromagnets with different polarities and takes advantage of a nature-inspired ratio, known as the golden ratio. In EFO, a possible solution is an electromagnetic particle made of electromagnets, and the number of electromagnets is determined by the number of variables of the optimization problem. EFO is a populationbased algorithm in which the population is divided into three fields (positive, negative, and neutral); attraction-repulsion forces among electromagnets of these three fields lead particles toward global minima. The golden ratio determines the ratio between attraction and repulsion forces to help particles converge quickly and effectively. The experimental results on 30 high dimensional CEC 2014 benchmarks reflect the superiority of EFO in terms of accuracy and convergence speed over other state-of-the-art optimization algorithms.

Fuzzy sign-aware diffusion models for influence maximization in signed social networks

Information Sciences, Oct 1, 2023

FSCN: A Novel Forwarding Method based on Shannon Entropy and COPRAS Decision Process in Named Data Networking

One of the next-generation Information-Centric Networking (ICN) architectures for the future Inte... more One of the next-generation Information-Centric Networking (ICN) architectures for the future Internet is Named Data Networking (NDN). The appropriate output interface (forwarding) has recently emerged as a key challenge affecting the performance of this network and has received a lot of attention from researchers due to its importance. Although efforts have been made through the methods introduced so far to address this challenge, the neglect of effective network load balancing parameters, non-dynamic weighting, and the use of only a few criteria have been suggested for selecting the output interface have made it vital to further improve the network delay, congestion, and ultimately the network's throughput. Regarding this matter, a new strategy is presented in this article based on the COPRAS Decision-Making Process with the dynamic weighting of Shannon's entropy called FSCN for transmitting request packets through the optimal output interface. The essential parameters iden...

Binary Accelerated Particle Swarm Algorithm (BAPSA) for discrete optimization problems

Journal of Global Optimization, Dec 14, 2012

Revolutionizing Learning Technology Based on Information Technology

The information society is a utopia of the scholars' target. To achieve information society k... more The information society is a utopia of the scholars' target. To achieve information society knowledge production and distribution is necessary and the starting point is education. Information technology and communication can develop basic steps involved in public education and promote community knowledge. One of the discussions considered by the advocaters of the development of information technology in education is that they can raise educational opportunities for different classes of society, especially the poor. Experience shows that the latter does not happen consistently and if the government does not finance for the access of the poor to resources, information technology gap which called "digital gap" increases daily. Our country needs to have a new attitude towards the education of the young eager to go to schools and universities, the uneducated, and the laborers in order to increase their efficiency in specific specialties and to improve training programs in s...

UTF: Upgrade transfer function for binary meta-heuristic algorithms

Applied Soft Computing, 2021

In the real world, many optimization problems are discrete and very complex to solve. Some of the... more In the real world, many optimization problems are discrete and very complex to solve. Some of them are in the class of NP-hard problems and their search spaces grow exponentially with the problem size. As a result, an exhaustive search will be impractical using exact algorithms. In the last decades, meta-heuristic algorithms as approximate algorithms have shown superior performance in solving these problems. The majority of these algorithms have been designed for continuous search spaces and are not able to solve binary optimization problems. Therefore, a transfer function is applied to convert the continuous search space to the binary one. The performance of such binary algorithms depends on their ability of exploration, exploitation and transfer function. Several transfer functions have been introduced so far but they have shown poor exploration and exploitation in solving some problems. In this study, a novel adaptive transfer function, based on two linear functions, is proposed to overcome the shortcomings of existing transfer functions. The proposed method called upgrade transfer function (UTF) adapts itself during running the algorithm to switch from exploration to exploitation. This capability also covers disadvantages of meta-heuristic algorithms in terms of poor exploration and exploitation. The performance of UTF has been evaluated by three discrete optimization problems: function optimization, feature selection and the 0–1 multi-knapsack problem (MKP). The results of binary particle swarm optimization (BPSO), binary artificial bee colony (BABC) and several improved BPSO and BABC have been compared with those of UTF–BPSO and UTF–BABC using function optimization problems. Also, the efficiency of UTF–BPSO and UTF–BABC and some binary meta-heuristic algorithms such as binary salp swarm algorithm (BSSA) and binary gray wolf optimization (bGWO), binary dragon algorithm (BDA), binary multi-neighborhood artificial bee colony (BMNABC), binary hybrid topology particle swarm optimization quadratic interpolation (BHTPSO-QI), binary ant lion optimizer (bALO) and binary gravitational search algorithm (BGSA) have been evaluated by feature selection problems. Moreover, UTF and seventeen transfer functions have been applied in original PSO, ABC, SSA and GWO algorithms to solve low and high dimensions 0–1 MKP benchmark instances. The results showed that the new transfer function significantly enhances the performance of algorithms to achieve the best solution in the binary search space.

Software Engineering and Information System Applications in life

In recent century, every country based on its political, social and cultural situation, has a spe... more In recent century, every country based on its political, social and cultural situation, has a special goal to follow in order to develop city management, in turn, to improve the quality of services without any limit. On the top of these services lie life standards and the presence of citizen in decision making. The best way to actualize this goal is the existence of cities with e-management. In traditional cities, citizens consume too much on useless repetitive activities. Banking, official, business, educational, scientific, recreational, and political affairs as well as health care are activities that a person may do every day. Due to the fact that most cities provided with emanagement, the citizen's access possibility to different facilities in offices and places is provided round-the-clock. The aim of this paper is to investigate e-city properties and the applications of software engineering and information technology in an e-city. In this regard, basic challenges are invest...

Improved centripetal accelerated particle swarm optimization

Particle swarm optimization (PSO) is a bio-inspired optimization algorithm that imitates the soci... more Particle swarm optimization (PSO) is a bio-inspired optimization algorithm that imitates the social behavior of bird flocking, fish schooling and swarm theory. Although PSO has a simple concept and is easy to implement, it may converge prematurely because of its poor exploration when solving complex multimodal problems. Centripetal accelerated particle swarm optimization (CAPSO) is an enhanced particle swarm optimization (PSO) scheme combined with Newton’s laws of motion. It has shown an effective algorithm in solving optimization problems however; its performance can be enhanced similar to other evolutionary computation algorithms. In this study, an improved CAPSO (ICAPSO) and improved binary CAPSO (IBCAPSO) are introduced to accelerate the learning procedure and convergence rate of optimization problems in the real and binary search spaces. The proposed algorithms are evaluated by twenty high-dimensional complex benchmark functions. The results showed that the methods substantiall...

R-GWO: Representative-based grey wolf optimizer for solving engineering problems

Applied Soft Computing, 2021

Abstract The grey wolf optimizer (GWO) is a well-known nature-inspired algorithm, which shows a s... more Abstract The grey wolf optimizer (GWO) is a well-known nature-inspired algorithm, which shows a sufficient performance for solving various optimization problems. However, it suffers from low exploration and population diversity because its optimization process is only based on the best three wolves greedily, and the information of other wolves does not consider. In this paper, a representative-based grey wolf optimizer (R-GWO) is proposed to tackle with these weaknesses of the GWO. The R-GWO introduces a search strategy named representative-based hunting (RH) a combination of three effective trial vectors inspired by alpha wolves’ behaviors to improve the exploration and diversity of the population. The RH search strategy utilizes a representative archive to reduces the greediness and enhance the diversity of solutions, and it can also strike balance between the exploration and exploitation using a non-linear control parameter. The performance and applicability of the proposed R-GWO were evaluated on CEC 2018 benchmark functions and six engineering design problems. The results were compared by eight state-of-the-art metaheuristic algorithms: PSO, KH, GWO, WOA, EEGWO, BOA, HHO, and HGSO. Moreover, the results were statistically analyzed by three test Wilcoxon rank-sum, Friedman and mean absolute error (MAE). The performance results show that on all 29 functions with dimensions 30, 50, and 100, the R-GWO is superior to the competitor algorithms except on function 27 on all dimensions and function 22 on dimension 30. The proposed R-GWO is the most effective algorithm compared with competitor algorithms, with an overall effectiveness of 95.4%. The experimental and statistical results show that the R-GWO is competitive and superior to compared algorithms and can solve engineering design problems better than competitor algorithms.

A novel x-shaped binary particle swarm optimization

Soft Computing, 2020

Definitive optimization algorithms are not able to solve high-dimensional optimization problems b... more Definitive optimization algorithms are not able to solve high-dimensional optimization problems because the search space grows exponentially with the problem size and an exhaustive search will be impractical. Therefore, approximate algorithms are applied to solve them. A category of approximate algorithms are meta-heuristic algorithms. They have shown an acceptable efficiency to solve these problems. Among them, particle swarm optimization (PSO) is one of the well-known swarm intelligence algorithms to optimize continuous problems. A transfer function is applied in this algorithm to convert the continuous search space to the binary one. The role of the transfer function in binary PSO (BPSO) is very important to enhance its performance. Several transfer functions have been proposed for BPSO such as S-shaped, V-shaped, linear and other transfer functions. However, BPSO algorithm can sometimes find local optima or show slow convergence speed in some problems because of using the velocity of PSO and these transfer functions. In this study, a novel transfer function called x-shaped BPSO (XBPSO) is proposed to increase exploration and exploitation of BPSO in the binary search space. The transfer function uses two functions and improved rules to generate a new binary solution. The proposed method has been run on 33 benchmark instances of the 0–1 multidimensional knapsack problem (MKP), two discrete maximization functions and 23 minimization functions. The results have been compared with some well-known BPSO and discrete meta-heuristic algorithms. The results showed that x-shaped transfer function considerably increased the solution accuracy and convergence speed in BPSO algorithm. The average error of compared algorithms on all 0–1 MKP benchmark instances indicated that XBPSO has the minimum error of 8.9%. Also, the mean absolute error (MAE) obtained by XBPSO on two discrete maximization functions is 0.45. Moreover, the proposed transfer function provides superior solutions in 18 functions from 23 minimization functions.

A new rainfall forecasting model using the CAPSO algorithm and an artificial neural network

Neural Computing and Applications, 2015

Artificial neural networks (ANNs) are being used increasingly to forecast rainfall. In this study... more

Improvement of Multi-Layer Perceptron (MLP) training using optimization algorithms

Evolutionary Algorithms for Neural Network Learning Enhancement

Artificial Neural Network (ANN) is one of the modern computational methods proposed to solve the ... more Artificial Neural Network (ANN) is one of the modern computational methods proposed to solve the majority of real world problems. Back-Propagation (BP) algorithm (as a gradient descend method) is one of the most popular methods for ANN training. However, there are some unavoidable disadvantages such as slow convergence speed, easily gets into partial extreme value and infirm global searching capability. To solve these problems, many solutions have been presented so far. Among them, Evolutionary Algorithms (EAs) have shown a good performance in this regard. EAs use some mechanisms inspired by biological evolution to find an optimization solution. Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Imperialist Competitive Algorithm (ICA) are in the class of algorithms. In this study, the mentioned optimization algorithms are chosen and applied in feed-forward neural network to enhance the learning process in terms of convergence rate and classification accuracy.

A review of population-based meta-heuristic algorithms

Enhancement of artificial neural network learning using centripetal accelerated particle swarm optimization for medical diseases diagnosis

Soft Computing, 2013

In recent decades, artificial neural networks (ANNs) have been extensively applied in different a... more In recent decades, artificial neural networks (ANNs) have been extensively applied in different areas such as engineering, medicine, business, education, manufacturing and so on. Nowadays, ANNs are as a hot research in medicine especially in the fields of medical disease diagnosis. To have a high efficiency in ANN, selection of an appropriate architecture and learning algorithm is very important. ANN learning is a complex task and an efficient learning algorithm has a significant role to enhance ANN performance. In this paper, a new meta-heuristic algorithm, centripetal accelerated particle swarm optimization (CAPSO), is applied to evolve the ANN learning and accuracy. The algorithm is based on an improved scheme of particle swarm algorithm and Newton's laws of motion. The hybrid learning of CAPSO and multi-layer perceptron (MLP) network, CAPSO-MLP, is used to classify the data of nine standard medical datasets of Hepatitis, Heart Disease, Pima Indian Diabetes, Wisconsin Prognostic Breast Cancer, Parkinson's disease, Echocardiogram, Liver Disorders, Laryngeal 1 and Acute Inflammations. The performance of CAPSO-MLP is compared with those of PSO, gravitational search algorithm and imperialist Communicated by D. Liu.

Binary Accelerated Particle Swarm Algorithm (BAPSA) for discrete optimization problems

Journal of Global Optimization, 2012

Your article is protected by copyright and all rights are held exclusively by Springer Science +B... more Your article is protected by copyright and all rights are held exclusively by Springer Science +Business Media New York. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com".

CAPSO: Centripetal accelerated particle swarm optimization

Information Sciences, 2014

ABSTRACT Meta-heuristic search algorithms are developed to solve optimization problems. Such algo... more ABSTRACT Meta-heuristic search algorithms are developed to solve optimization problems. Such algorithms are appropriate for global searches because of their global exploration and local exploitation abilities. Swarm intelligence (SI) algorithms comprise a branch of meta-heuristic algorithms that imitate the behavior of insects, birds, fishes, and other natural phenomena to find solutions for complex optimization problems. In this study, an improved particle swarm optimization (PSO) scheme combined with Newton’s laws of motion, the centripetal accelerated particle swarm optimization (CAPSO) scheme, is introduced. CAPSO accelerates the learning and convergence of optimization problems. In addition, the binary mode of the proposed algorithm, binary centripetal accelerated particle swarm optimization (BCAPSO), is introduced for binary search spaces. These algorithms are evaluated using nonlinear benchmark functions, and the results are compared with the gravitational search algorithm (GSA) and PSO in both the real and the binary search spaces. Moreover, the performance of CAPSO in solving the functions is compared with some well-known PSO algorithms in the literature. The experimental results showed that the proposed methods enhance the performance of PSO in terms of convergence speed, solution accuracy and global optimality.

A Review of Emotional Learning And It's Utilization in Control Engineering

Int. J. Advance. Soft Comput. Appl, 2010

In mammalians, brain Limbic System is responsible for emotional process. An important part of mot... more