napomena: {x} se odnosi na pasus, na temu x, kojega ce Drazen napisati:)
------------------------------------------------------------------------

draft#0:

A Survey of Mutation Approaches
in Genetic Algorithms for Internet Search

DD+BN+VM

Abstract {one sentence per section, after all is done}


1. Introduction

In the introductory section we specify the following three issues of importance: (a) Definition of the general field of interest for this research, which is genetic search on the Internet, (b) Definition of a critical issue of interest for this field, which is mutation approach for genetic search, and (c) Definition of the viewpoint of interest for this survey (definition of the axiomatic requirement of this survey), which is optimization of price/performance in the Internet environment. This is the first paper in our series of survey papers on critical issues in genetic search on the Internet (the follow up papers are related to: presentation of genomes, crossover, off-spring fitness factor calculation, selection, and application to indexing for browsing purposes).

{genetic search - 7Ws}

{mutation algorithms for genetic search - 7Ws}

{Internet search and how to measure price/performance - 7Ws}


2. Problem statement

In this section we focus on the following issues of importance for the understanding of the basic orientation of this survey: (a) Problem definition, which is to define, classify, and explain, to both, novices eager to learn, and experts in the field eager to get updated, which means that the structure of this survey is "onion-like," (b) Elaboration of problem importance (why is it important to clasify and elaborate in this field; here we draw a parallel with the Mendelyeyev system in chemistry), and (c) An assesment of the problem development trends (discussing why will the importance of this problem grow over time).

A major issue of importance for our survey is to create a taxonomy of algorithm for mutation in genetic search on the Internet and in general, using three different binary criteria, which implies the formation of 8 different classification categories or for short, classes. As it will be seen later, two of the formed classes do not include examples, although something like that would make sense. This scenario opens up two more avenues for furhter research, and that is where, conditionally speaking, we see an analogy with the Mendelyeyev system in chemistry: The empty boxes of the Mendelyeyev classification defined the features of the elements not know at the time of classification formation. The fact that some classification boxes were empty, was a stimulus for researchers to focus into a given set of directions, which resulted in new inventions (discoveries of new chemical elements). Likewise, the empty classes of our classification do open up new research focuses in the area of mutation for genetic algorithms.

{why is the problem important}

{why will the importance of this problem grow}


3. Existing surveys and their criticism

In this section, we give a short overview of existing survey paper in genetic search in general, and in the domain of mutation approachs, in specific. For each one we define the 7Ws (when, where, why, who, whom, whom, and how), the major characteristic, high points, drawbacks, and we define the major differences in comparison with our survey (six long sentences per survey effort).

{survey#1}

{survey#N}


4. Classification Criteria and the Classification Tree

The major two problems in creation of a new taxonomy are: the classification criteria and the classification tree. Here, the classification criteria were chosen to reflect the essence of the basic viewpoint of this research. The classification tree was obtained by successive application of the chosen criteria. The leaves of the classification tree are the examples (research efforts) elaborated briefly later on, in the Existing Solutions section of this paper.

After the criteria and the classes are defined and explaind, this surevey continues with an overview which, for each and every example (selected research effort), gives the following six main points: (a) The 7 Ws of the research (again: who, where, when, whom, why, what, how), (b) Essence of the approach (the main strategic issue which differ from the research presented before), (c) Details of the approach (the main tactical issues of interest for the research presented here), (d) Further development trends of the approach, and (e) A criticism of the approach (looking from the viewpoint of our research), and finally (f) Possible improvements that could overcome the noticed drawbacks (which is what the future research of others can benefit from).

For space limitation reasons, each one of the above six points is given a sentence (is given a page in our follow-up book for a major USA publisher). For bette runderstanding reasons, each example is also described with a figure, using the terms and shapes identical to those in all other figures of this survey paper (by including a figure per example, we support the opinion that one figure is worth one thousand words; by insisting on identical terms and shapes, we enable that different examples be compared using a "common denominator").

The classification criteria of interest for this research, as well as the justification there-off, are given in Table 1. All selected classification criteria are briefly explained in the caption of Table 1, and elaborated in the paragraphs to follow.

USING A MUTATIONAL DATABASE OR NOT /criterion#1
USING SYSTEM BASES OF INTERNET PROVIDERS OR NOT /criterion#2
USING APPLICATION BASES OF END USERS OR NOT /criterion#3

Table 1: Classification criteria.
Legend: {one definition per in-figure block}.
Explanation: {one strong sentence}.

{one paragraph for each criterion, i.e. three paragraphs}

The classification tree, derived from the above introduced classification criteria, is presented in Figure 1, and has eight leaves. Each leaf of the classification tree is given a name, as described in Table 2.

DRAWING OF A TREE
Figure !: Classification Tree and the List of Classes
Legend: {}
Explanation: {}

{one paragraph for each class, i.e. eight paragraphs}

TABLE OF SYMBOLIC NAMES (NAMES OF GREEK GODS)
Table 2: Symbolic names of classes

For each leaf (class), the list of existing solutions (examples) is given in a separate paragraph - just the names of approaches and major references, to enable interested readers to study further details, after the essence becomes clear after reading this survey. If a leaf contains no known solutions, but does make sense, the paragraph explains why it makes sense to focus future research into this direction. If a leaf contains no known solutions, because such a class does not make sense, it is explained why such a combination of features makes no sense.

{1: one paragraph list of solutions under each leaf}
{2:}
{8:}


5. Presentation of existing solutions

This section is divided in several subsections, one per leaf of the above defined classification. For each leaf, as already indicated, several 6-sentence paragraphs are given, one per research effort overviewed. In the text to follow, instead of the terms leaf or class, we use the term Solution Group.

5.1. SolutionGroup#1[...]: {MPU?}

{one 6-sentence-paragraph and one figure per solution}

5.2. SolutionGrop#2: {}

{}

5.3. SolutionGroup#3: {}

{}

5.X. SolutionGroup#4: {}

{}


5.X+1. Conclusion about existing solutions: local conclusion

From all above presented, we conclude, among the existing solutions, the one which can be treated as the best one, for the general axiomatic viewpoit of this research is {}. In the analyis part of a follow-up paper, the best existing solution is used as the counterpart against which the advantages of our proposed solution, introduced by [], and presened here in brief as a possible solution for an empty class.


6. The trend and reminescences about optimal solutions for future

{the application view-point}

{the technology view-point}

Figure 2: The essence of a proposed generalized approach.
Legend: {}. Explanations: {}.

{essence, elaborated}

{adventages, elaborated}


7. Summary: Global conclusion

{what was done, recapitulated}
{to whom is this of importance}
{what are the newly open research problems}
{something wise for the end, for readers to remember forever}

8. Acknowledgements

{}

9. References (annotated bibliography)

9.1. Genetic algoriths

9.2. Survey articles

9.2. Mutational algorithms

9.4. Miscelaneous issues