Talk:Artificial bee colony algorithm
From Scholarpedia
REVIEWER A
This Scholoarpedia article summarizes the development of Artificial Bee Colony algorithm. It is an interesting article with very good description of the algorithm and its applications.
However, there are some important issues/points which needs to be addressed in the revision:
1) There are some strong similarities between Artificial Bee Colony (ABC) algorithm (described in this article) and Bee Colony Optimization of wikipedia (see Bee colony optimization, http://en.wikipedia.org/wiki/Bee_colony_optimization ) They both talk about the algorithm inspired by the foraging behaviour of Honey Bees. So the difference and similarity should be explained.
2) The description of the pseudo code is too complicated. Ideally, it should be in a few steps in a box with a frame, and then description how each steps work. Then, give an example. In addition, the description of the ABC algorithm for constrained optimization involves 26 steps. This is very unusually lengthy. We know the author try to give as much detail as possible. But the description is not easy to follow.
3) There is no comparison with other algorithms. It is difficult to see what are advantages other than a new algorithm? Why people should use this algorithm? Under what conditions? The references should be more complete.
REVIEWER B
The article requires major revisions. These are listed below:
1) The overall structure is confusing: (i) Why do you start with a section on Swarm Intelligence? A Scholarpedia entry on Swarm Intelligence is already available; (ii) You should start the paper with a short, authoritative definition of what the "artificial bee colony" algorithm is. (iii) Then continue with an example. (iv) Then give a precise, formal definition of the algorithm and of the class of problems it applies to. (v) Give general information on the existing applications and on current research. (vi) Finally, you might add an appendix on the natural inspiration.
You can use the cited "ant colony optimization" entry in scholarpedia as a good example of the structure you could use for your entry.
2) English has to be improved.
3) I agree with the comments given by REVIEWER A.
REVIEWER B -- SECOND REVIEW, ABOUT REVISED VERSION
The article still requires major revisions:
1) The author did not implement all points of my previous point 1). In particular, points (ii) and (v) were not implemented. Also the suggestion to use the cited "ant colony optimization" entry in scholarpedia as a good example was not followed. Because of this, the paper still present major problems. I try to list them in the following:
- The entry starts without a definition of what the ABC algorithm is about. As a reader, I want an initial short definition.
- Section 1 is overly complicate. However, the main problem with this section is that it is not clear at all why we do need all this stuff to explain the algorithm. All the details used (e.g., in figures 1 to 4) are never used in the paper. So, this section should be removed as it confusing. A shorter version could be added as appendix. In the appendix should be given only the information necessary to understand why the algorithm is called "Artificial Bee Colony algorithm". All the unnecessary details should be removed.
- Section 2 should be much improved. Some of the problems I identified:
(i) Terminology problem:
You say: "... the position of a food source represents a possible solution ..."
Then: ""The population of solutions $x_{ij}$ are initialized ..."
Then: "Each employed bee determines a food source ($v_{ij}$), ..."
Now, I do not understand the difference beteen $x_{ij}$ and $v_{ij}$
You also say: "After producing a new solution ($v_i$) ..."
Wasn't $v_i$ a food source?
If $v_i$ is a solution, why a single index and not two (i.e., $v_{ij}$)?
(ii) Another terminology problem:
You write: "The population of solutions $x_{ij}$ are initialized in the range of the parameter $j$."
Then you give equation (1) saying that $xmin_j$ and $xmax_j$ are lower and upper bounds of parameter $j$.
I am sorry, but I really do not understand.
$x_{ij}$ has two indexes, $i$ and $j$. Why do you consider only $j$? What is $i$?
- Section 3:
(i) The table caption should say what are the cells entries
(ii) "From the results presented in Table 1, it is clear that ABC algorithm is more successful
than PSO at training neural networks on the XOR benchmark problem in three cases."
This sentence has not much meaning given we neither know which PSO algorithm you used, nor
we know if this PSO algorithm (or PSO algorithms in general) are good at this problem.
- Section 4 should be improved. I will check it in details at the next round of review.
- Section 5 is far too detailed for an encyclopedic entry.
- Section 6 should be improved.
- A section discussing all the "Bee colony" algorithms available in the literature is necessary so as to avoid confusion.
- The References section should be much improved in formatting:
(i) There are wrong accents (ii) Why Germany and TURKEY?? (iii) On The Performance Of -> On the Performance of (iv) MANY other small problems
- See also: A link to the Swarm Intelligence entry is in order.
REVIEWER B -- THIRD REVIEW, ABOUT RE-REVISED VERSION
The entry remains in my opinion quite confusing. The main problem is that, with the information given, it is impossible to reproduce the algorithm. In fact, it is even impossible to understand it.
Other important problems are: the writing is often confusing, with terms that are either not properly explained, or different terms used to refer to the same thing, math notation that is not uniform across the entry, sentences with typos and English errors, and so on.
I give below a number of comments, but the authors should make a serious effort to improve the paper beyond what I indicate.
Artificial Bee Colony algorithm -> The Artificial Bee Colony algorithm ABC algorithm -> The ABC algorithm Change everywhere
However, "ABC mimics" is OK without "the".
Artificial Bee Colony (ABC) algorithm, … , simulates the intelligent foraging behavior of honey bee swarms. This is not precise and misleading. The goal of your algorithm is not to simulate the intelligent foraging behavior. Rather, you take inspiration from the intelligent foraging behavior of honey bee swarms to write your optimization algorithm.
The minimal model of forage selection … ABC algorithm mimics this minimal model of optimization process … So, is a model of forage selection or of optimization process?
three essential components: food sources, … behaviour: recruitment to a … You cannot have twice the colon sign in a same sentence.
feedback(Tereshko -> feedback (Tereshko … add the missing space. There are plenty of these small problems. The entry cannot be accepted until they are ALL removed.
ie. -> i.e.,
… ie. the food sources with rich nectar and also close to the hive. Bad sentence. Please rephrase.
The meaning of inequalities (2) should be given just after explaining (1) and before explaining (3).
) and domains of variables are limited by their lower and upper bounds (2). -> ). The variables domains are limited by their lower and upper bounds (2). Move above.
bees: employed bees associated with a specific food source, and unemployed bees: onlookers -> You cannot have twice the colon sign in a same sentence.
to decide a site -> to choose a site
The number of the employed bees is equal to the number of food sources, each of which also represents a site, being exploited at the moment or to the number of solutions in the population. Rephrase in better English.
In ABC optimization, the steps given below are repeated until a stopping criteria is satisfied: -> The general scheme of the ABC algorithm is as follows:
is an alternative solution vector -> is a solution vector
Why don't you use the same notation as the one introduced in (1) to (4)? There a solution (parameter vector) is x, and its components i go from one to n. In the description of the algorithm solutions become x_i, and parameters go from 1 to D rather than n.
Employed bees search for new food sources (v_i), which are is also representative of different sites … What is the difference between "food sources" and "site"? They look the same to me. If this is the case, then why to use two terms for a same thing? Otherwise, please explain better.
the probability values, p_i with that x_i is preferred by onlookers, are calculated for the solutions … This part belongs to the Onlooker Bees Phase, it seems to me.
In equation (8) why f_i rather than f(x_i)?
which are is also -> another example of typo
the probability values, p_i with that x_i is preferred by onlookers … Rephrase in better English.
onlookers -> onlooker bees
… the nectar amount of a food source corresponds to the quality (fitness) of the associated solution. As mentioned before, the nectar amount of a food source corresponds to the quality of the solution represented by that food source position. For example, if the nectar of x_i has been exhausted … How can you exhaust the quality of a solution?? Also, the way the “nectar” value changes over time is never explained.
Why to use two names for the same type of bees: scouts and explorers? It is confusing.
The rest of the paper should also be improved … I will give detailed comments once all the problems mentioned above have been taken care of.
Current trend -> Current trends
Use alphabetic order for references.
REVIEWER B -- FOURTH REVIEW, ABOUT RE-RE-REVISED VERSION
The entry is still not acceptable and I wonder if it will ever be ... :
- the author did not provide an answer to my comments in review 3; - most of the comments I did in review 3 still apply; - I have implemented a few corrections and indicated a few of the many problems directly in the text, using capital letters; - there are far too many references, and their style is still neither uniform nor correct; - finally, the most important point is that the entry is very difficult to read, it is not authoritative and requires serious additional work before it can be accepted: the author is kindly asked to avoid submitting a new version until it has taken serious care of all my comments; also, he should let a few of his colleagues (among which it would be good to add a native speaker) read the entry and he should collect their comments; I suggest that researchers who are not familiar at all with ABC should read the entry and provide additional comments.
My personal judgement is that the entry is no closer to be publishable than it was at the time of its initial submission and that, unless the author makes a serious effort, it will not reach a level at which I can accept it.
REVIEWER B -- FIFTH REVIEW
This time I have made a number of low-level corrections directly in the text. In the text I have also added comments in CAPITAL letters. These comments need to be addressed before the paper can be accepted. Please make the changes and provide an answer to my comment.
REVIEWER C
Development of Artificial Bee Colony algorithm is summarized in this Scholoarpedia article. This interesting article is described in very good description of the algorithm and its applications with figures and flows.
Even if the article explains well the ABC algorithm, there are some points to be improved.
1) The focus of article should be on ABC algorithm to explain it simple as it is.
2) It is better to compare the algorithms with other to show its simplicity.
Response to Reviewer A
1- Bee Colony Optimization (BCO) was developed by Teodorovic for combinatorial type problems. ABC algorithm was developed by Karaboga for numerical optimization. Algorithms are based on different models. Altough the title is Bee Colony Optimization, the content in wikipedia does not belong to the essential BCO, since the wikipedia contents are not moderated. It belongs to Bees Algorithm proposed by Pham et al., If you mean Bees Algorithm(BA) by BCO, the differences between the BA and ABC can be found in the paper recently presented and discussed at IPROMS'09 conference which is a virtual conference at link: http://conference.iproms.org/artificial_bee_colony_abc_harmony_search_and_bees_algorithms_ on_numerical_optimization 2- Description was simplified. 3- In the applications, the results of PSO and \mu+\lambda-ES algorithms were added.
Response to Reviewer B
1- The description and the explanations were re-structured. 2- English was improved. 3- Please refer to the respose to Reviewer A.
Response to Reviewer C
1- Explanation was simplified. 2- Comparisions with PSO and \mu+\lambda-ES algorithms were added.
Second Response to Reviewer B
- the article starts with a short definition telling the user what the ABC algorithm is.
-In the "Currrent Trend" section, a general information on the existing applications and current research are stated.
- foraging behaviour of bees are explained in Appendix section and figures are removed.
-Section 2 has been improved and the problems identified by the reviewer are corrected. In initialization phase, what x_i, x_{ij},j, xmin_j are described.
- The caption of Table 1 is changed telling what are the cell entries.
- PSO is changed as standard PSO and the reference giving details about this experiment has been added.
-Section 4 is removed in order to avoid confusion. Instead, how ABC is applied to constrained opt. problem is stated in Section "A Constrained Engineering Problem: Welded Beam Design" and the reference is given for details. (Karaboga D., Basturk B. (2007), Artificial bee colony (ABC) optimization algorithm for solving constrained optimization problems, LNCS: Advances in Soft Computing: Foundations of Fuzzy Logic and Soft Computing, Vol: 4529/2007, pp: 789-798, Springer- Verlag, 2007, IFSA 2007.)
-The details of the section "A Constrained Engineering Problem: Welded Beam Design" is removed.
- a section is added for other approaches simulating honeybee foraging behaviour
-References are improved in formatting
Response to Fourth Revision of Reviewer B
First, we would like to thank to the Reviewer for his/her invaluable suggestions.
Reviewer B: The model consists of three essential components: food sources, employed and unemployed foraging bees. Foraging bees search for rich food sources close to their hive. WHAT ABOUT THE TWO OTHER COMPONENTS??
A. Second sentence was rewritten as the following:
The model consists of three essential components: food sources, employed and unemployed foraging bees. The two components, both employed and unemployed foraging bees, search for rich food sources, which is the third component, close to their hive.
Reviewer B: In ABC, a colony of artificial forager bees (agents) search rich artificial food sources (good solutions for a given problem). In the application WHAT APPLICATION?, first, the optimization problem is converted into the problem of finding the best parameter vector which minimizes an objective function.
A. “In the application” expression was replaced with “In real-life optimization problems where the ABC algorithm is to be used”.
In real-life optimization problems where the ABC algorithm is to be used, the optimization problem is first converted to the problem of finding the best parameter vector which minimizes an objective function.
Reviewer B: Secondly, artificial bees discover a population of initial solution vectors and then iteratively improve them by employing the strategies: recruitment to good solutions, abandonment of poor solutions; and using a neighbour production mechanism. SHOULD BE REPHRASED.
A.Sentence was modified as given below.
Second, the artificial bees randomly discover a population of initial solution vectors and then iteratively improve them by employing the strategies: moving towards better solutions by means of a neighbour search mechanism while abandoning poor solutions.
Reviewer B: where 
is defined on a search space, 
, which is a 
-dimensional rectangle in 
(
) REPHRASE: BAD ENGLISH. The variable domains are limited by their lower and upper bounds (11).
A. Problem definition was rephrased as given below:
A global optimization problem can be defined as finding the parameter vector 
that minimizes an objective function, , as below:
- (1)
which is constrained by the following inequalities and/or equalities:
- (11)
- (12)
- (13)
is defined on a search space, , which is a 
dimensional rectangle in 
(
). The variable domains are limited by their lower and upper bounds (11).
- (12)
- (13)
THE "for" IN THE ABOVE FORMULA SHOULD BE IN STANDARD FONT AND NOT IN MATH NOTATION.
A. “for” was written in standard font.
- (11)
- (12)
- (13)
Reviewer B: In ABC, the first half of the colony consists of employed bees while the second half constitutes the onlooker bees. WHAT I UNDERSTAND HERE IS THAT THERE ARE NO SCOUTS
A. In order to make the classification of forager bees more clearly, this sentence was replaced with the following paragraph:
Both onlookers and scouts are also called unemployed bees. Initially, all food source positions are discovered by scout bees. Thereafter, the food sources are exploited by employed bees and onlooker bees. In case the food source of an employed bee is exhausted, then that employed bee becomes a scout bee in search of further food sources ones again. In other words, the employed bee whose food source has been exhausted becomes a scout bee.
Reviewer B: The number of employed bees is equal to the number of food sources (solutions) since each employed bee is associated with ONE AND only one food source.
A. Sentence was corrected as suggested.
Reviewer B: === Initialization Phase ===
The population of food sources 
IS initialized 
, 
the population size) "m" WAS ALREADY USED IN EREVİEWER B:4! .
A. Eq. 4 was rewritten as
- (11)
- (12)
- (13)
Reviewer B: Since each food source, 
, is a solution vector to the optimization problem, each vector holds elements 
that correspond to the parameters WHY PARAMETERS AND NOT VARIABLES? IN EReviewer B: 1 TO 4 YOU WERE TALKING OF VARIABLES. HERE AND IN THE FOLLOWING OF PARAMETERS. BE CONSISTENT. of an optimization problem and is the number of optimization parameters.
A. In Eq. 4, m was changed as q and the sentence was rewritten as the following:
Since each food source, , is a solution vector to the optimization problem, each vector holds number of variables, (), which are to be optimized so as to minimize the objective function.
Reviewer B: After producing THE new food source 
, its fitness is calculated and a greedy selection is applied between 
and .
A. Sentence was corrected as suggested.
Reviewer B: The fitness value might be calculated for minimization problems using the following FORMULA (14)
A. Sentence was corrected as suggested.
Reviewer B: AS BEFORE: "if" SHOULD BE IN TEXT FONT.
A. Mathematical notation of “if” was corrected
- (14)
Reviewer B: ALSO WHAT IS "fit"??
A. A definition for fitness was written:
The fitness value of the solution, 
, might be calculated for minimization problems using the following formula (14)
Reviewer B: where 
is the objective function VALUE OF solution 
.
A. where is the objective function value of solution .
Reviewer B: The probability value 
with WHICH is chosen by an onlooker bee can be calculated by using the expression given in equation () :
A. Sentence was corrected as suggested:
The probability value with which is chosen by an onlooker bee can be calculated by using the expression given in equation () :
Reviewer B: As in THE employed bees phase, a greedy selection is applied between and .
A. Sentence was corrected as suggested:
As in the employed bees phase, a greedy selection is applied between and .
Reviewer B: The number of trials for abandoning a solution is an important control parameter called ‘‘limit’’ or “abandonment criterion”. THIS SENTENCE SHOULD BE REWRITTEN.
A. Sentence was rewritten as below:
Employed bees whose solutions cannot be improved through a predetermined number of trials, specified by the user of the ABC algorithm and called “limit” or “abandonment criteria” herein, become scouts and their solutions are abandoned. Then, the contraverted scouts start to search for new solutions, randomly.
Reviewer B:
Main features of the ABC algorithm ===
The ABC algorithm
1) mimics the foraging behaviour of honeybees closely -- NOT TRUE
2) has been originally proposed for numerical optimization -- NOT A FEATURE OF THE ALGORITHM!
3) can be also used for combinatorial optimization problems -- NOT A FEATURE OF THE ALGORITHM!
4) can be used for unconstrained and constrained optimization problems -- NOT A FEATURE OF THE ALGORITHM!
5) has the ability of local and global search -- NOT CLEAR TO ME WHAT IS THE MEANING OF THIS
6) is quite simple, flexible and robust -- NOT SUPPORTED BY DATA (ESPECIALLY THE FLEXIBLE AND ROBUST PARTS)
7) employs a few control parameters -- SENTENCE IS TOO GENERIC
A. References were added for the claims and this paragraph was reorganized.
In summary, the ABC algorithm,
1) attemps to mimic the foraging behaviour of honeybees.
2) is a global optimization algorithm.
3) has been initially proposed for numerical optimization (e.g.: Karaboga, 2005).
4) can be also used for combinatorial optimization problems (eg: Pan et al, 2010).
5) can be used for unconstrained and constrained optimization problems (eg: Karaboga and Akay, 2009);Karaboga07b|Karaboga and Basturk 2007b]]; Domínguez 2009).
6) employs only three control parameters (population size, maximum cycle number and limit) that are to be predetermined by the user.
7) is quite simple, flexible and robust (Some of relevant publications expressing these merits of the ABC algorithm are Rao et al, 2008;Kang et al, 2009; Singh, 2009; Karaboga, 2009 included in the References list).
Reviewer B: The exclusive-OR (XOR) Boolean function is a difficult classification problem mapping two binary inputs to a single binary output as (0 0;0 1;1 0;1 1)›(0;1;1;0). This classical benchmark is also a hard task for the neural networks that are being successfully applied to solving problems in pattern classification, function approximation, optimization, pattern matching and associative memories. Training an artificial neural network is an optimization task since it is desired to find the optimal weight set of a neural network in training process. The optimization goal is to minimize the objective function OBJ..FUN. NOT DEFINED! by optimizing the network weights. THIS WHOLE PAR SHOULD BE REWRITTEN.
A. This paragraph was rewritten as suggested:
Training an artificial neural network is an optimization task since it is desired to find the optimal set of weights of a neural network in the training process. The goal is to optimize the network weights by minimizing an objective function such as mean square error (MSE) given by (15).
- (15)
.
where, 
is the error at the 
th iteration; 
, the weights in the connections at the th iteration; 
, the desired output node; 
, the actual value of the 
th output node; 
, the number of output nodes; , the number of patterns.
The neural networks are being successfully applied to solving problems in pattern classification, function approximation, optimization, pattern matching and associative memories.
The exclusive-OR (XOR) XOR is a difficult classification problem mapping two binary inputs to a single binary output as (0 0;0 1;1 0;1 1) (0;1;1;0). This classical benchmark problem is a hard task also for the neural networks.
Reviewer B: In Table 1, means of Mean Square Errors (MSE) of 30 runs of each configuration are recorded for ABC and Particle Swarm Optimization (PSO) algorithms WHICH EXACT PSO WAS USED?
A. This paragraph was changed as below:
In Table 1, mean MSE values of 30 runs of each configuration are recorded for ABC and standard Particle Swarm Optimization (PSO) Eberhart and Kennedy, 1995; algorithms where each run started with a random population with different seeds.
Reviewer B: === A Constrained Optimization Problem: Welded Beam Design===
YOU SHOULD START WITH A CONCISE PROBLEM DEFINITION.
In order to handle with constraints in the constrained optimization problem addressed, the ABC algorithm employed Deb’s rules. Deb’s rules are used instead of the greedy selection employed between and in ABC proposed for unconstrained optimization problems (Karaboga and Basturk, 2007). Deb’s method uses a tournament selection operator, where two solutions are compared at a time by applying following criteria (Deb, 2000):
A. The section was started by the definition of welded beam problem and reorganized as the following:
A Constrained Optimization Problem: Welded Beam Design===
The welded beam design is a common real-life application problem frequently encountered in steel structures. As illustrated in Fig.2, the problem is dimensioning a welded steel beam and the welding length so as to minimize its cost subjected to constraints on shear stress, 
, bending stress in the beam, 
, buckling load on the bar, 
, end deflection of the beam,
, and side constraints. There are four design variables: 
, which in structural engineering are commonly symbolized by the letters shown in Fig. 2 (
). Structural analysis of this beam leads to the following nonlinear objective function subjected to five nonlinear and two linear inequality constraints as given below.
- (16)
The optimum solution is located on the boundaries of the feasible region, and the ratio of the feasible region to the entire search space is quite small for this problem, which makes it a truly difficult problem for any optimization algorithm.
Generally, a constraint handling technique should be incorporated to the optimization algorithms proposed for solving unconstrained problems. Therefore, in order to handle the constraints of this problem, the ABC algorithm employs Deb’s rules, which are used instead of the greedy selection employed between and in ABC proposed for unconstrained optimization problems (Karaboga and Basturk, 2007). Deb’s method uses a tournament selection operator, where two solutions are compared at a time by applying the following criteria (Deb, 2000):
Reviewer B: Numerical Optimization:
Yang (2005) developed a virtual bee algorithm (VBA) to solve the numerical function optimizations WHAT IS "the numerical function optimizations"?? .
A. “Numerical function optimization” was changed to “numerical optimization problems”:
In 2005, Yang (2005) developed a virtual bee algorithm (VBA) to solve numerical optimization problems. The original algorithm works with only two variables. Pham et al. (2005) described the Bees Algorithm which performs a kind of neighbourhood search combined with random search and can be used for both combinatorial optimization and numerical optimizations.
Reviewer B: The use of ABC for optimizing hybrid functions, solving integer programming and engineering design problems (Rao et al., 2008,Singh, 2009, Karaboga, 2009), combinatorial type optimization, multi-objective optimization and binary-variable optimization are current RESEARCH topics.
A. This sentence was changed as suggested:
The use of ABC for optimizing hybrid functions, solving integer programming, and engineering design problems (Rao et al., 2008; Singh, 2009; Karaboga, 2009), combinatorial type optimization (Pan et al, (2010)), multi-objective optimization (Omkar et al, (2010)), clustering (Karaboga and Ozturk, (2010)) and binary-variable optimization are current research topics.
Reviewer B: == References ==
THE NUMBER OF REFERENCES SHOULD GREATLY BE REDUCED. REFERENCES SHOULD ALL USE THE SAME SYLE: Behav Ecol Sociobiol =/= Comput. Inf. Syst. J. =/= Journal of Global Optimization : CHOOSE A COMMON STYLE
A. References were reformatted.
Reviewer B: ==Appendix -- Foraging Behaviour of Honey Bees== ….
THE FOLLOWING THREE POINTS ARE A REPETITION.
(i) Food sources: The “profitability” of a food source is related to several factors such as its closeness to the nest, richness of energy and the ease of extracting the energy from the source. In the minimal model, the profitability of a food source is represented with one of the quantities.
(ii) Employed foragers: These foragers are associated with a specific food source they exploit or are employed at. They carry information about the specific source such as its distance, direction from the hive and the profitability of the source and then they share this information with the forager bees waiting in the hive by dancing which is an example of multiple interaction.
(iii) Unemployed foragers seeking a food source to exploit: scouts and onlooker bees . The scouts randomly search the environment surrounding the hive for new food sources and this behaviour is a kind of fluctuations which is vital for self-organization; and the onlooker bees waiting in the hive find a food source by means of the information presented by employed foragers. The mean number of scouts is about 5–10% of the foragers (Tereshko and Loengarov, 2005)
A. These components were rewritten as below to avoid repetition problem:
(i) Food sources: The “profitability” of a food source is related to several factors such as its closeness to the nest, richness of energy, and the ease of extracting the energy from the source. In the minimal model, the “profitability” may be defined by one of these quantities.
(ii) Employed foragers: These foragers are associated with a specific food source they exploit. They carry information to the hive and share it with other foragers waiting in the hive by dancing.
(iii) Unemployed foragers: These foragers consist of scouts and onlookers. The scouts randomly search the environment surrounding the hive for new food sources, and the onlooker bees waiting in the hive detect a food source by means of the information presented to them by the employed foragers (Tereshko and Loengarov, 2005).
The exchange of information among the foragers is very important for the formation of collective knowledge. The most important part of the hive for exchanging information is the dancing area where different types of dances are performed: Waggle dance, Round dance, Tremble dance, etc. Communication among bees related to the quality of food sources is called the waggle dance. Since information about all the current rich sources is available to onlooker bees on the dance floor, they watch numerous dances and direct themselves to profitable sources. Employed foragers share their information in proportion to the profitability of their food sources. As the information circulating about them increases, the probability of the onlooker bees choosing the more profitable sources also increases (Tereshko and Loengarov, 2005).
Response to Fifth Revision of Reviewer B
First, we would like to thank to the Reviewer for his/her invaluable suggestions.
Reviewer B: This problem is also known as constrained optimization problem. If 
, then it is called unconstrained optimization problem. WHAT ABOUT 
?
A. This sentence was rewritten as the following:
If it is an unconstrained optimization problem, then both 
and .
Reviewer B: Thereafter, the food sources are exploited by employed bees and onlooker bees. In case the food source of an employed bee is exhausted IT IS NOT CLEAR HOW A FOOD SOURCE CAN BECOME EXHAUSTED., then that employed bee becomes a scout bee in search of further food sources once again.
A. These sentences were modified as the following:
Thereafter, the nectars of food sources are exploited by employed bees and onlooker bees, and this continual exploitation will ultimately cause them to be exhausted. Then, the employed bee which was exploiting the exhausted food source becomes a scout bee in search of further food sources once again.
Reviewer B: . Then, the contraverted ?? NOT AN ENGLISH WORD ... scouts start to search for new solutions, randomly
A. Sorry about this. The word “converted” had been miswritten as “contraverted”. This word was corrected.
Then, the converted scouts start to search for new solutions, randomly.
Reviewer B: Hence those sources which are initially poor and/or become poor by exploitation IT IS NOT CLEAR HOW A SOURCE CAN BECOME POOR. THIS SHOULD BE EXPLAINED
A. This sentence was rewritten as the following:
Hence those sources which are initially poor or have been made poor by exploitation are abandoned and negative feedback behaviour arises to balance the positive feedback.
Reviewer B: The exclusive-OR (XOR) XOR is a difficult classification problem mapping two binary inputs to a single binary output as (0 0;0 1;1 0;1 1) WHAT IS THIS SYMBOL? (0;1;1;0). This classical benchmark problem is a hard task also for the neural Networks
A. The symbol was modified as “>” The exclusive-OR (XOR) XOR is a difficult classification problem mapping two binary inputs to a single binary output as (0 0;0 1;1 0;1 1)> (0;1;1;0). This classical benchmark problem is a hard task also for the neural networks.
Reviewer B: THE FORMATTING OF THE FOLLOWING PARAGRAPH SHOULD BE CORRECTED:
Any feasible solution satisfying all constraints is preferred to any infeasible solution violating any of the constraints,
Among two feasible solutions, the one having better fitness value is preferred,
Among two infeasible solutions, the one having the smaller constraint violation is preferred.
A. This paragraph was reformatted.
Reviwer B: THE FOLLOWING PARAGRAPHS SHOULD BE REMOVED AS YOU JUST STATE THAT THERE WERE COMPARSIONS WITHOUT SAYING WHAT WAS THE RESULT. ALSO, IT WOULD TAE FAR TOO MUCH SPACE TO DESCRIBE THE RESULTS AND THIS SHOULD NOT BE DONE IN AN ENCYCLOPEDIC ENTRY.
The performance of ABC was compared against the various previous algorithms such as: Genetic Algorithm (GA), Differential Evolution (DE), and Particle Swarm Optimization (PSO) for unconstrained numerical optimization problems (Karaboga and Basturk, 2007a; Karaboga and Basturk, 2008; Karaboga and Akay, 2009).
An extended version of ABC was compared on some constrained optimization problems with DE and PSO algorithms in (Karaboga and Basturk 2007b, Domínguez 2009).
The ABC algorithm was applied to train feed-forward neural networks on classification problems. For example, in (Karaboga et al, 2007), XOR, Decoder-Encoder and 3-Bit Parity benchmark problems were chosen to compare the performance of ABC with Genetic Algorithm (GA) and Back-Propagation (BP) Algorithms. In ( Karaboga and Ozturk, 2009), medical problems were considered from Proben1 database (Prechelt, 1994) and the classification performance of the ABC algorithm was compared to BP, Levenberg-Marquardt (LM) and GA, DE and PSO.
A. These paragraphs above were removed as suggested by the reviewer.
Reviewer B: References
THIS SECTION SHOULD BE IMPROVED. REMOVE ALL REFERENCES THAT ARE NO LONGER CITED IN TEXT. ADD CITY AND COUNTRY OF PUBLISHERS. MAKE SURE JOURNAL NAMES ARE GIVEN. IN GENERAL, MAKE SURE ALL CITATIONS ARE COMPLETE
All citations that are no longer cited in the text were removed and Publishers and countries of the journals in the references were provided.
