# Artificial Bee Colony Algorithm

## Top PDF Artificial Bee Colony Algorithm:

### New Local Search Strategy in Artificial Bee Colony Algorithm

be in the range of [-1, 1]. The value of limit (maximum cycle number) should be SN × D, where, SN is the number of possible solutions and D is the dimension of the problem. Wei-feng Gao et al. [11] proposed an improved solution search method in ABC, which depends on the fact that bee searches around the best solution of the preceding iteration to increase the exploitation. A. Banharnsakun et al. [12] introduced a new variant of ABC namely the best-so-far selection in artificial bee colony algorithm. To enhance the exploitation and exploration processes, they propose to make three major changes by introducing the best-so-far method, an adjustable search radius, and an objective-value- based comparison method in DE. J.C. Bansal et al. [13] anticipated balanced ABC; they added a new control parameter, Cognitive Learning Factor and also tailored range of ɸ in Artificial Bee Colony algorithm. Qingxian and Haijun [14] anticipated a change in the initialization scheme by making the initial group symmetrical, and the Boltzmann selection mechanism was employed instead of roulette wheel selection for humanizing the convergence ability of the ABC algorithm.

### Automatic Image Enhancement by Artificial Bee Colony Algorithm

Recently, similar to the existing nature inspired algorithms, a new mimic algorithm inspired by the behaviors of bees, named Artificial Bee Colony algorithm, was proposed by Karaboga and Basturk [8]. In [9], it has been shown that ABC algorithm is superior to other optimization algorithms such as GA, PSO and Differential Evolution (DE), etc. Due to the simplicity and robustness of ABC algorithm, it has been implemented to solve various problems in image processing, robot path planning, parameter identification, job-shop scheduling.

### IMPLEMENTATION OF PARALLEL ARTIFICIAL BEE COLONY ALGORITHM ON VEHICLE ROUTING PROBLEM

Artificial Bee Colony algorithm has proved to be useful in various fields of problems such as assignment problems, scheduling problems, solving transportation problems etc. In this paper we will particularly apply Artificial Bee Colony algorithm to solve vehicle routing problem efficiently. We have also tried to obtain better quality of results by opting for very large search spaces Moreover we will speed up execution of ABC algorithm. For this purpose we have opted for the parallelization of ABC algorithm.

### Optimizing the Path Traversed using Artificial Bee Colony Algorithm

In this paper, we have highlighted the need of path optimization. Due to increasing time constraints, there is a need to cover the path in the most optimum manner. Where, the path is defined as the trajectory to be followed starting from a specific start position up to a predefined destination or goal point. Further, we have described Artificial Bee Colony Algorithm, which is one of the many techniques of Swarm Intelligence and can be successfully applied for optimizing the path. Moreover, a comparison between two Swarm Intelligence techniques, namely Artificial Bee Colony algorithm and Ant Colony Optimization algorithm is presented. It is observed that ants are good in search and exploitation and thus ACO can be used for dynamic applications. However, its theoretical analysis is difficult and the probability distribution changes by iteration. Moreover, the time for convergence is uncertain. ABC on the other hand, employees fewer parameters, it has strong robustness, fast convergence and high flexibility. It can also be used for solving multimodal and multidimensional optimization problems. In addition to this, ABC has global optimization and easy recognition. It conducts both local search and global search in each iteration and as a result the probability of finding the optimal increases. The structure of the ABC algorithm is such that it supports parallel processing as result saving time. Considering the wide number of advantages of Artificial Bee Colony algorithm (ABC) in comparison to other Swarm intelligence techniques, it can be concluded that ABC is the most efficient algorithm for optimizing a given path.

### Research on Application of Artificial Bee Colony Algorithm in Facial Expression Recognition

Emoticon classification is an important step in facial expression recognition. In order to improve the recognition rate. Ensemble Learning method use multiple classifiers to determine the expression category. However, among those classifiers in Ensemble Learning, there are some classifiers that are redundancy and poor performance. In order to solve this problem, Artificial Bee Colony algorithm is used to assign different weights to different classifiers, which gives high weight to the classifier performance better, otherwise, it gives low or zero weight to the classifier performance bad. So as to optimize the selection. Experimental results show that this method not only improves the expression recognition rate, but also reduces the computational cost.

### An Artificial Bee Colony Algorithm to Mine Periodic High Utility Itemsets

From the experimental results, it clearly shows the ABC based algorithms performs better than the state of art PHUIM algorithms. High utility itemset mining based on artificial bee colony algorithm is proposed. The proposed PHUIM-ABC algorithm mines 50% faster than the state-of-art algorithms. The candidate itemset generated by the proposed PHUIM-ABC has 97% correct value compared to the FHM algorithm. For mushroom data set it performs 50% faster for all the values of minimum utility threshold. The memory usage is reduced by a minimum of 40% for the mushroom, connect and retail datasets. For accident and foodmart the memory usage is reduced by minimum of 20%. Since retail industries need a very fast output than the exact outputs; this algorithm will be best suited. The correctness of the algorithm is above 80 % for all the dataset and for all the minimum threshold value. In the future it is planned to implement genetic algorithms for PHUIM to improve efficiency. This algorithm can be also extended to sequential utility mining algorithms.

### An Improved Artificial Bee Colony Algorithm with Elite-Guided Search Equations

Recently, Cui et al. [5] proposed an artificial bee colony algorithm (the ABC elite) with two novel search equations. One search equation incorporates the beneficial infor- mation of elite solutions, which is applied to the employed bee phase, the other one not only exploits the valuable information of the elite solutions, but also employs that of the current best solution used in the onlooker bee phase. Furthermore, the ABC elite is embedded into depth-first framework to form a new variant of ABC, the DFSABC elite. Experimental results show that ABC-elite and DFSABC elite are very effective compared with other recently proposed ABC variants.

### An Improved Memetic Search in Artificial Bee Colony Algorithm

value of limit (utmost cycle number) should be SN × D, where, SN is the number of solutions and D is the dimension of the problem. W Gao et al. [17] anticipated an enhanced solution search equation in ABC, which is based on the fact that bee searches only around the best solution of the previous iteration to increase the exploitation and introduce a selective probability. A. Banharnsakun et al. [18] proposed a novel variant of ABC that is to say the best-so- far selection in artificial bee colony algorithm. In this algorithm the best possible solutions established so far are shared globally in the midst of the entire population. Thus, the new contender solutions are more plausible to be close to the in progress best solution. In other words, we bias the solution direction toward the best-so-far position. Moreover, every succession adjusts the radius of the search for new individual using a larger radius previously in the search procedure and then reduces the radius as the process comes closer to converging. Finally, it uses a more robust calculation to determine and put side by side the quality of alternative solutions. To enhance the exploitation and exploration processes, they propose to make three major changes by introducing the best-so-far method, an adjustable search radius, and an objective-value-based comparison method in DE. J.C. Bansal et al. [56] wished- for balanced ABC; they introduced a novel control parameter, Cognitive Learning Factor and also modified range of ɸ ij in Artificial Bee Colony algorithm.

### Optimization of power systems through an artificial bee colony algorithm

The integration of optimization methods in the different processes involved in an electric power system in the search for energetic efficiency has generated satisfying results in the reduction of energy consumption, technical losses, increasing security and system reliability. The purpose of this article is to implement the artificial bee colony optimization algorithm in a 15-node IEEE power system set at 13.2 kV, in order to find the possible values of the reactive compensation that optimize the system power flow. In first place, the results of the voltage profiles of a 15-node IEEE power distribution system are shown with the Newton Raphson method. Then, said system is optimized using an adapted version of the artificial bee colony algorithm which was developed in MATLAB. After the execution of the algorithm, it was concluded that the nodal voltage values have a significant increase in all 15 nodes of the system. This translates into a reduction of the losses in the interconnection lines of the nodes through the optimization of the power system. The application of the artificial bee colony algorithm offers an optimization alternative driven to reduce the energy losses in the power system. Keywords: algorithm, optimization, power flow, power system.

### Analysing convergence, consistency and trajectory of Artificial Bee Colony Algorithm

I N the recent past, researchers have shown interest in algorithms inspired from natural phenomena. To name a few we have Particle Swarm Optimisation (PSO) algorithm [23] taking inspiration from birds flocking, Artificial Bee Colony (ABC) optimization algorithm [21] inspired by foraging behaviour of honey bees, Gravitational Search Algorithm (GSA) [28] taking inspiration from law of gravity and interaction between the masses, Harmony Search Algorithm (HSA) [12] inspired by improvisation done by jazz musician, Differential Evolution (DE) algorithm [30] inspired by theory of evolution and Spider Monkey Optimization (SMO) [4] algorithm taking inspiration from foraging behaviour of spider monkeys. Recently various variants of ABC algorithm have been proposed which includes modified global best artificial bee colony for constrained optimization problem [3], artificial bee colony algorithm with multiple search strategies [11], an adaptive artificial bee colony algorithm for global optimization [35], hybrid artificial bee colony with differential evolution [18][34], simulated annealing based artificial bee colony algorithm for global numerical optimization [6] and escalated convergent artificial bee colony [17]. Study has shown that these algorithms are considered as an efficient solver of complex optimization problems. Artificial Bee Colony (ABC) optimization algorithm and its variants has been applied to various optimization problems such as solving partition and scheduling problem in codesign [24][14], artificial neural networks [25], forecasting stock markets [15], automatic software fault localization [16], parameter identification for Van Der Pol - Duffing oscillator [10], network topology design [29] and structural engineering [8].

### A non revisiting artificial bee colony algorithm for phased array synthesis

It is observed that a great number of function eval- uations is needed for artificial bee colony algorithm to obtain a satisfactory solution. Accordingly, a great num- ber of cycles or generations is evolved during optimization process. It is well known that population diversity in EAs and SI algorithms heavily reduces once algorithms are evolved in a large number of generations. This means solutions tend to be homogeneous or most genetic infor- mation becomes identical. Thus, such algorithms could not produce diverse variations and then lose the capabil- ity of refining fitness of solutions. In ABC, scout bees is responsible to diversify solutions for having a wider search area. However, recent study shows that this method is not effective to speed up convergence rate of ABC [27]. In this paper, a non-revisiting scheme is used to keep population diversity substituting for scout bee stage in ABC. The idea of non-revisiting scheme is to restrict an algorithm from revisiting an already searched place. For one thing, it avoids revisiting and repeated solution evaluation. For another thing, it memorizes all visited places by algorithm so as to let the search of algorithm focus on unknown places with high uncertainty. Besides non-revisiting scheme, three bee groups in ABC are also changed. As non-revisiting scheme is apt to guide the search directions of an algorithm, employed and scout bees are eliminated from standard ABC algorithm. Hence, only onlooker bee stage is kept in the resulting algo- rithm. The new algorithm is named as non-revisiting artificial bee colony (NrABC). NrABC is then applied to tackle phased array design problems. Numerical experi- ment is conducted studying NrABC and standard ABC. The results are discussed and analyzed at the end of the paper.

### A MODIFICATION OF ARTIFICIAL BEE COLONY ALGORITHM FOR SOLVING INITIAL VALUE PROBLEMS

function, we trained the network both traditional Artificial Bee Colony algorithm and a variant of ABC proposed by us. Proposed algorithm uses dynamically constructed hy- persphere to generate new bee population. The individuals in new population fall into the hypersphere to increase the exploitation ability of traditional ABC. Similarly, the individuals generated outside of the hypersphere supports the exploration quality of ABC. In this work, we give some numerical examples some different types of differential equa- tions such as first order and second order ODEs. The empirical studies precisely clarify that the modified version of ABC outperforms the classical ABC by means of absolute and mean squared errors. Table 6 exposes that cost values obtained in training and the testing stages are quite similar. Only, the desired improvement has not been achieved for the second order differential equation. Furthermore, it can be observed that the improvement has been drastic at the initial steps of the algorithm with Figure 2. However, it has been slight for following steps of the proposed algorithm.

### Improved Artificial Bee Colony Algorithm for Continuous Optimization Problems

Since invention of ABC 2005, studies on ABC in the literature have increased significantly. The ABC was used for designing of digital IRR filters by Karaboga [7], Singh used it for leaf-constrained minimum spanning tree problem [8], Rao et al. proposed ABC for optimization of distribution network configuration for loss reduc- tion [9]. The ABC was implemented to solve quadratic minimum spanning tree problem by Sundar and Singh [10]. A modified ABC for real parameter optimization was proposed by Akay and Karaboga [11]. Karaboga and Akay modified ABC for solving constrained optimization problems by using Deb’s rules [12]. Pan et al. devel- oped a discrete model of ABC for lot-streaming ﬂow shop scheduling problem [13]. The ABC was also used for solving reliability redundancy allocation problems [14], neural networks training [15], software test suite opti- mization [16]. In addition, Gbest-guided ABC for numerical function optimization was proposed by Zhu and Kwong [17] and Alatas proposed a chaotic artificial bee colony algorithm for avoiding to get stuck on local so- lutions [18].

### ARTIFICIAL BEE COLONY ALGORITHM FOR LOCALIZATION IN WIRELESS SENSOR NETWORKS

Thus, it is concluded that MAP-ABC algorithm minimizes the localization error better than MAP-M&N. Hybridization of optimization namely Simulated Annealing (SA) can be combined to Mobile Anchor Positioning with Artificial Bee Colony algorithm (MAP-ABC-SA) so as to reduce the localization error further and the localization error of the hybrid evolutionary algorithm can be compared with the pure ABC algorithm (MAP-ABC) to validate its performance. The future enhancement may be applying meta-heuristic optimization approaches such as Glow worm swarm optimization, fish swarm optimization etc. with mobile anchor positioning to further minimize the localization error significantly in wireless sensor networks.

### An enhanced hybridized artificial bee colony algorithm for optimization problems

Recently, many swarm-bsed algorithms have been proposed, including genetic algorithm (GA) [1], particle swarm optimization algorithm (PSO) [2], ant colony optimization algorithm (ACO) [3], differential evolution algorithm (DE) [4], harmony search algorithm (HS) [5], artificial bee colony algorithm (ABC) [6]. ABC proposed by Karaboga is one of the most popular swarm-based algorithms, which is based on the intelligent foraging behavior of honey bee swarm. For the reason that it is simple and easy to implement, ABC algorithm has attracted a lot of scholars’ attention. Benchmark functions experiment has shown that ABC is competitive over GA, DE and PSO algorithm. Since proposed, ABC has been widely used in optimization problems.

### Path Based Test Suite Augmentation using Artificial Bee Colony Algorithm

Artificial Bee colony algorithm (ABC) [10] has found its usage in various fields of software testing like Test suite Optimization, Automated Generation of Pair wise Tests. This research utilizes the behavior of bee in order to find all the independent paths in the modified program. The proposed algorithm is used for identifying the affected paths in a program. The bee also performs a comparison between the path list of original and modified programs and based on this comparison affected paths are selected. The existing test suite is run on these affected paths to check its adequacy. If the original test suite is not sufficient enough to handle changes then new test cases are generated in order to achieve 100% coverage. The algorithm has been executed on 8 examples. It is able to detect the paths that have been affected by changes. The results have shown 100% path coverage and the generation of optimal test suite that can handle changes effectively.

### Artificial Bee Colony Algorithm for Optimizaion in Data Science

Abstract— Data science is all about performing various operations on various type of data. Big data is a large amount of data which is hard to handle by on hand systems. It requires new structures, algorithms and techniques. As data increases as per volume, dark data also will increase. Artificial Bee Colony algorithm is a part of Swarm Intelligence. It is based on how honey bees are working to find out their food sources. In Big Data there is distributed environment so required sources may be on different places. During process the data these data sources have to find out from different places and analyze a one system. This requires calculation which can help us to find out best option for our required data sources. ABC algorithm is used to overcome limitations of ant colony algorithm. In ant colony initialization will be repeat from starting point in case of failure. In bee colony optimization initialization happens only once. It is used to find out required data source based on parameters out of multiple data sources. Thus, artificial bee colony algorithm can be used to find out best data sources. We can store these derived data sources on cloud for further processing. Bee colony algorithm generally used in data mining and networking field. It can be used for Big Data for identifying data resources.

### A Comprehensive review of Artificial Bee Colony Algorithm

[Chidambaram and Lopes (2009)] proposed a new method which applied the Artificial Bee Colony Algorithm (ABC) to recognize objects in the images. The objective was to find a pattern or reference image (template) of an object somewhere in a target landscape scene. Considering that it may be translated, scaled, rotated and/or partially occluded to identify th e given reference image in the target landscape image. The best solutions obtained in their experiments with gray scale and color images indicated that ABC algorithm was much faster than the other evolutionary algorithms and with comparable accuracy. They also claimed that to the best of their knowledge, this was the first application of the ABC algorithm to this sort of problem.