Swarm intelligence (SI) is briefly defined as the collective behaviour of decentralized and self-organized swarms. The well known examples for these swarms are bird flocks, fish schools and the colony of social insects such as termites, ants and bees. In 1990s, especially two approaches based on ant colony and on fish schooling/bird flocking introduced have highly attracted the interest of researchers. Although the self-organization features are required by SI are strongly and clearly seen in honey bee colonies, unfortunately the researchers have recently started to be interested in the behaviour of these swarm systems to describe new intelligent approaches, especially from the beginning of 2000s. During a decade, several algorithms have been developed depending on different intelligent behaviours of honey bee swarms. Among those, artificial bee colony (ABC) is the one which has been most widely studied on and applied to solve the real world problems, so far. Day by day the number of researchers being interested in ABC algorithm increases rapidly. This work presents a comprehensive survey of the advances with ABC and its applications. It is hoped that this survey would be very beneficial for the researchers studying on SI, particularly ABC algorithm.

A comprehensive survey: artificial bee colony (ABC) algorithm and applications

This paper reviews studies using empirical, data-driven methodologies, such as cluster analysis and latent class analysis, to identify clustering patterns of diet, physical activity and sedentary behavior among children or adolescents and their associations with socio-demographic indicators, and overweight and obesity.

The clustering of diet, physical activity and sedentary behavior in children and adolescents: a review

Fundamentals of statistical signal processing: Estimation theory: by Steven M. KAY; Prentice Hall signal processing series; Prentice Hall; Englewood Cliffs, NJ, USA; 1993; xii + 595 pp.; $65; ISBN: 0-13-345711-7

Artificial bee colony (ABC) algorithm inspired by the foraging behaviour of the honey bees is one of the most popular swarm intelligence based optimization techniques. Quick artificial bee colony (qABC) is a new version of ABC algorithm which models the behaviour of onlooker bees more accurately and improves the performance of standard ABC in terms of local search ability. In this study, the qABC method is described and its performance is analysed depending on the neighbourhood radius, on a set of benchmark problems. And also some analyses about the effect of the parameter limit and colony size on qABC optimization are carried out. Moreover, the performance of qABC is compared with the state of art algorithms' performances.

A quick artificial bee colony (qABC) algorithm and its performance on optimization problems

A survey on applications and variants of the cuckoo search algorithm

https://www.techscience.com/cmc/v70n2/44706/pdf

Training Multi-Layer Perceptron with Enhanced Brain Storm Optimization Metaheuristics

The salp swarm algorithm is one of the novel swarm intelligence metaheuristics. The work proposed in this paper provides further improvements of the salp swarm algorithm, that have been achieved by modifications of the original approach. By analyzing solutions’ quality and convergence speed of basic salp swarm during practical simulations, it was concluded that the exploitation process can be improved. Improvements were achieved by introducing the concept of opposite solutions in the initialization phase, as well as in iterative search process, where a fine-tuned exploitation of the current best solution is performed by generating its opposite individual. Proposed improved salp swarm algorithm was tested on thirteen well-known global benchmarks. Comparative analysis was performed with seven other modern metaheuristics methods, and against the original salp swarm algorithm. Accomplished results have proven that proposed approach in a large degree outscores original algorithm and other approaches included in comparative analysis.

Current Best Opposition-Based Learning Salp Swarm Algorithm for Global Numerical Optimization

Feature selection problem from the domain of machine learning refers to selecting only those features from the high dimensional datasets, that have prominent influence on dependent variable(s). In this way, dataset dimensionallity is reduced and only the riches data is kept, training process of machine learning model becomes more efficient and accuracy is increased. This manuscript proposes a new hybridized version of the sine cosine algorithm adjusting for solving feature selection problem. Hybridization is relatively novel approach for combing and improving metaheuristics optimizer. Notwithstanding that the basic sine cosine algorithm establishes good performance for solving NP hard challenges, based on simulation results, it was concluded that there is still space for improvement in its exploitation process. Original sine cosine algorithm and proposed hybridized implementation were tested on a well-known 21 machine learning datasets retrieved from the UCL repository. Comparative analysis between hybrid sine cosine and original one, as well as with 10 other state-of-the-art metaheuristics was conducted. Established results in terms of classification accuracy and fitness prove the robustness and efficiency of proposed method for solving this type of NP hard challenge.

Feature Selection in Machine Learning by Hybrid Sine Cosine Metaheuristics

Higher level of image processing usually contains some kind of recognition. Digit recognition is common in applications and handwritten digit recognition is an important subfield. Handwritten digits are characterized by large variations so template matching, in general, is not very efficient. In this paper we describe an algorithm for handwritten digit recognition based on projections histograms. Classification is facilitated by carefully tuned 45 support vector machines (SVM) using One Against One strategy. Our proposed algorithm was tested on standard benchmark images from MNIST database and it achieved remarkable global accuracy of 99.05%, with possibilities for further improvement.

An algorithm for handwritten digit recognition using projection histograms and SVM classifier

Cloud computing technology refers to on-demand access to services, applications, and infrastructure that runs on a distributed network utilizing virtualized resources. In the cloud model, an efficient task scheduling algorithm plays an important role in order to achieve better functioning in general and resource utilization of the cloud. The end-users submit tasks, and the scheduling algorithm needs to allocate them to the available resources on time. Task scheduling issue is considered as NP-hard problems, and metaheuristics algorithms demonstrate high efficiency in solving such problems, thus, in this work, we propose enhanced flower pollination algorithm for the task scheduling. The major focus of this study is to reduce the makespan. We compared the results of the proposed method to other similar approaches, such as PBACO, ACO, Min-Min, and FCFS allocation strategies. The obtained results from the experiment show that the proposed EEFPA scheduler has the potential to allocate submitted tasks by the user to the available resources on the cloud.

Nebojsa Bacanin

Papers

Training Multi-Layer Perceptron with Enhanced Brain Storm Optimization Metaheuristics

Current Best Opposition-Based Learning Salp Swarm Algorithm for Global Numerical Optimization

Feature Selection in Machine Learning by Hybrid Sine Cosine Metaheuristics

An algorithm for handwritten digit recognition using projection histograms and SVM classifier

Enhanced Flower Pollination Algorithm for Task Scheduling in Cloud Computing Environment