Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning

System

Name of Student - Harshal Jain (10103538)

Name of Supervisor - Ms. Mukta Goel

A Genetic Algorithm (or GA) is a search technique used in computing to find true or approximate solutions to optimization and search problems. Genetic algorithms are categorized as global search heuristics. GA are a particular class of evolutionary algorithms that use techniques inspired by evolutionary biology such as inheritance, mutation, selection, and crossover (also called recombination).The most common type of genetic algorithm works like this: A population is created with a group of individuals created randomly. The individuals in the population are then evaluated. The evaluation function is provided by the programmer and gives

the individuals a score based on how well they perform at the given task.

Two individuals are then selected based on their fitness, the higher the fitness, higher the chance of being selected.

These individuals then "reproduce" to create one or more offspring, after which the offspring are mutated randomly.

This continues until a suitable solution has been found or a certain number of generations have passed, depending on the needs of the programmer.

Initialization

Initially many individual solutions are randomly generated to form an initial population. The population size depends on the nature of the problem, but typically contains several hundreds or thousands of possible solutions. Traditionally, the population is generated randomly, covering the entire range of possible solutions (the search space).

Occasionally, the solutions may be "seeded" in areas where optimal solutions are likely to be found.

Selection

During each successive generation, a proportion of the existing population is selected to breed a new generation. Individual solutions are selected through a fitness-based process, where fitter solutions (as measured by a fitness function) are typically more likely to be selected.

Certain selection methods rate the fitness of each solution and preferentially select the best solutions. Other methods rate only a random sample of the population, as this process may be very time-consuming.

Most functions are stochastic and designed so that a small proportion of less fit solutions are selected. This helps keep the diversity of the population large, preventing premature convergence on poor solutions. Popular and well-studied selection methods include roulette wheel selection and tournament selection.

In roulette wheel selection, individuals are given a probability of being selected that is directly proportionate to their fitness.

Two individuals are then chosen randomly based on these probabilities and produce offspring.

CrossoverThe most common type is single point crossover. In single point crossover, you choose a locus at which you swap the remaining alleles from on parent to the other. This is complex and is best understood visually. As you can see, the children take one section of the chromosome from each parent. The point at which the chromosome is broken depends on the randomly selected crossover point. This particular method is called single point crossover because only one crossover point exists. Sometimes only child 1 or child 2 is created, but oftentimes both offspring are created and put into the new population. Crossover does not always occur, however. Sometimes, based on a set probability, no crossover occurs and the parents are copied directly to the new population. The probability of crossover occurring is usually 60% to 70%.

MutationAfter selection and crossover, you now have a new population full of individuals. Some are directly copied, and others are produced by crossover. In order to ensure that the individuals are not all exactly the same, you allow for a small chance of mutation. You loop through all the alleles of all the individuals, and if that allele is selected for mutation, you can either change it by a small amount or replace it with a new value. The probability of mutation is usually between 1 and 2 tenths of a percent. Mutation is fairly simple. You just change the selected alleles based on what you feel is necessary and move on. Mutation is, however, vital to ensuring genetic diversity within the population.

TerminationThis generational process is repeated until a termination condition has been reached. Common terminating conditions are:

A solution is found that satisfies minimum criteria Fixed number of generations reached Allocated budget (computation time/money) reached The highest ranking solution's fitness is reaching or has reached a plateau such that successive iterations no longer produce better results Manual inspection Any Combinations of the above

Problem StatementThe aim of this project is to analyze the parameter, for the inputs to find an optimization problem than the candidate solution we have. This will help us to find more accurate knowledge level of user, using Genetic Algorithm (GA). In this algorithm a population of candidate solutions (called individuals, creatures, or phenotypes) to an optimization problem is evolved toward better solutions.

Solution ApproachThe solution approach is to find an optimized approach of knowledge level of user by finding the appropriate weight to power up the input value. This will tell us that whether the knowledge level is very low, low, medium, or high.

Support for Novelty

Here, the problem signifies that we have a candidate solution which is a member of a set of possible solutions to a given problem (a candidate solution does not have to be a likely or reasonable solution to the problem—it is simply in the set that satisfies all constraints; that is, it is in the set of feasible solutions.). So, in a genetic algorithm, a population of candidate solutions (called individuals, creatures, or phenotypes) to an optimization problem is evolved toward better solutions. Each candidate solution has a set of properties (its chromosomes or genotype) which can be mutated and altered; traditionally, solutions are represented in binary as strings of 0s and 1s, but other encodings are also possible.

Product PerspectiveSystem interfaces: The genetic algorithms ensures efficient output and subsequent visual access for ease-of-use.User interfaces: The integrated graphs formed the output values shows the knowledge level and make it easy to decide for every user. The designed genetic algorithm allows the user to see the parameter to use for finding the knowledge level it, then proceeds to yield the output on the screen.Hardware interfaces: The system does not have any hardware requirements.Software interfaces: As the system is coded upon eclipse SDK, the system must require eclipse.Title: Eclipse Classic 4.2.2 (32-bit)Author: The Eclipse Foundation (www.eclipse.org)Communications interfaces: The system does not require any network tools or interfaces.Memory: A minimum of 180 MB is required by the eclipse SDK. The device must also have adequate memory for training data, test data and code files.Operations: The user needs to have the degrees of the user as an input to find the parameter and user will be given the final result thereafter for the further decisions.Site adaptation requirements: The file must be compatible and size must not be greater than 30 MB. 

Overall Description:

Product FunctionsTake the degrees of the usersRandomly generate a variable form 0.1 to 0.9 which will used as to generate the power the degrees. There are 5 inputs in each row now the generated value is ‘w’ and input is ‘p’ now it finds p^w, where for each and every p there is distinct value of w.Then the selection process will select two chromosomes which will used for crossover and generates the 2 children of those two parents. The cross over will occur for 80% of population.After that in mutation we will randomly change the value of any ‘w’ which will lead us to termination.

User CharacteristicsAny entity can be a user if they have the degree or anything for which they have to find out the appropriate solution. 

ConstraintsReliability requirements: The algorithm designed is reliable for the user though it sometimes may deviate only a little from the original but then also it can be considered for the desired output.  Assumption and DependenciesBasic computer skills are assumed on the part of user.The user must have basic knowledge of Mysql.There must be a fully appropriate input, the inputs must not be hypothetical or assumed only by a small survey. Apportioning of RequirementsOnly one selection method is yet to be considered totally appropriate which may help finding the solution to this types of problem.

 

Functional Requirements:There is no authentication to any user because it is a simple code which requires a candidate solution to perform its feature.In this project there is a huge use of random values. There are 7 to 8 random values generated in the code according to their use.User Interface is simply the output of the project which shows the user an optimized way to its solution.Data to be used as input cannot be manipulated because a small difference in any value which is the part of the input will change the whole solution of the project.

Non-Functional Requirements: Reliability - The system shall be able to provide a level of

precision. System Requirements – The system requires Eclipse

software, php designer, wampp server. Database - The database must be regularly updated. Availability – In case of a crash, the system must be

restarted. Portability – Java, the language the algorithm employs, is

platform independent and php is platform dependent. There is a need of external interface for the algorithm.

ImplementationTake the degrees of the users in the database.Randomly generate a variable form 0.1 to 0.9 which will used as to generate the power the degrees. There are 5 inputs in each row now the generated value is ‘w’ and input is ‘p’ now it finds p^w, where for each and every p there is distinct value of w.It forms chromosomes of the population.Then we use the fitness function to find the fitness values of the chromosomes.Then the selection process will select two chromosomes which will be used for crossover and generates the 2 children of those two parents. The cross over will occur for 80% of population.After that in mutation we will randomly change the value of any ‘w’ which will lead us to termination.

FindingsThe following are the key findings during the course of this project:The main finding was that the selection process has been used two times – 1) to choose randomly any two chromosomes from population and 2) to choose the chromosomes to be used in the population after crossover by fitness function.It is sometimes difficult to find the right parameter to get our output, it may take many generations to find that which would take time.

ConclusionHere we conclude that the optimized solution of the candidate solution may come in the starting generations only or it will take many generations to come to that solution which is time consuming. But we will definitely find more accurate solution which will help us to achieve our aim and using this Genetic Algorithm for it.

Future WorkIn the future work this project will be very useful to find the predicted values according to the field in which this project can be helpful.Genetic Algorithm implementation cannot be easily found in java language for future which may become helpful to anyone needed in specific language. Genetic algorithms can be applied to a wide range of problems which are NP (Nondeterministic ally Polynomial) complete.GA future work is most useful in travelling salesperson problem, where a salesperson must find the shortest route through a number of cities starting and ending at a base city. However more practical applications include strategy planning, scheduling / time tabling and machine learning.

ReferencesPublished Papers:[1] “Marketing a web-site using a fuzzy logic approach” by IOAN CONSTANTIN ENACHE [2] “Ten years of genetic fuzzy systems: current framework and new trends” by O. Cordon, F. Gomide, F. Herrera, F. Hoffmann, L. Magdalena [3] Tuning Of Fuzzy Systems Using Genetic Algorithms by Ulrich Bodenhofer [4] Solving an aggregate production planning problem by using multi-objective genetic algorithm - (MOGA) approach by Md. A. Akhtar Hasin

ReferencesWeb Links:[1] http://www.wseas.us/e-library/conferences/2010/Bucharest/RIMA/RIMA-59.pdf[2]http://www.researchgate.net/publication/220529867_Ten_years_of_genetic_fuzzy_systems_current_framework_and_new_trends/file/32bfe50cb15ab289f2.pdf[3]http://www.researchgate.net/publication/200048700_Tuning_Of_Fuzzy_Systems_Using_Genetic_Algorithms/file/79e4150bde4e2da834.pdf [4] http://www.growingscience.com/ijiec/Vol4/IJIEC_2012_64.pdf