Abstract
Today, in the age of artificial intelligence and machine learning, Data mining and Image processing are two important platforms. GA and GP are value based and program based randomized searching tools respectively and these two are very much useful in the fields' data mining and image processing for handling different issues. In this chapter, a review is made on ability of GA and GP in some applications of these two fields. Here, the selected subfields of data mining are market analysis, fraud detection, risk management, sports analysis, protein interaction, classification of data, drug discovery and feature construction. The similar in image processing are enhancement and segmentation of images, face recognition, photo mosaic generation, data embedding, image pattern classification, object detection and Graphics Processor Unit (GPU) development. The efficiencies of GA and GP in these particular applications are analyzed with corresponding parameters, comparing with other non-GA and non-GP approaches of the corresponding subfields.
TopIntroduction
In present decade, optimization is an essential part in our everyday life. In any type of efficiency analysis, this is an important in each platform. Optimization procedures are majorly divided into two parts- hard computing and soft computing. Soft computing is more realistic than hard computing in every situation. Genetic Algorithm (GA) is a useful optimization approach of soft computing.
Genetic Algorithm (GA) was invented by John Holland in mid-sixties in university of Michigan. But idea of natural selection and natural genetics were developed by David Goldbarg. This is probabilistic and global search depending on principles of evolution. It was built up on the basis of three different hereditary properties of animals- selection, cross-over and mutation. Actually, the concept of creation of children from previous generation is used to obtain global optimum value in this useful as well as interesting soft computing approach.
According to differentiation of the operators’ selection, cross-over and mutation, different GA schemes are developed. According to number of objectives, there are two type genetic algorithms are made up like each type optimization approach- single objective genetic algorithms and multi objective genetic algorithms. Binary coded GA, Real coded GA, Messy GA etc are some single objective GA approaches and weighted sum, NSGA and NSGA-II etc are some multi objective GA approaches. In different platforms, different GA shows their optimization efficiencies in different ways. Among them, computer science, biology, economics, mathematics, pharmacology etc. are major application fields of genetic algorithm. In computer science, some efficient fields of GA are artificial intelligence, network security, data mining, image processing, pattern recognition etc.
Genetic Programming (GP) is an extended version of GA. In GA, population is created with a set of values and in GP; population is created with a set of sub programs, more specifically with a set of instructions. The concept of GP was first proposed by Lawrence J. Fogel in 1964. But at that time, he only introduced about this idea and simply implemented it in finite state automata. It was broadly used in 1990 with machine learning and artificial intelligence. Some specific application fields of GP are expert system, horn clause, networking, automation theory, software engineering, data mining, image processing etc.
In this chapter, we mainly focus on the application of GA and GP in data mining and image processing. We choose these two fields of application, because- these two are in a broad range of particular research application as well as GA and GP both are used in these two particular cases. So, we can analyze the efficiency of GA and GP with proper experiments.
Key Terms in this Chapter
Genotype: Full hereditary information of organism, used in GA and GP.
Mutation: Genetic operator for maintaining genetic diversity of population, i.e. changes on genes from one generation to another.
Phenotype: Actual observed properties of an organism, used in GA and GP.
Artificial Intelligence: Intelligence of machine or machinery tools to do particular task perfectly and quickly.
Crossover: Interchanging one (or more) portion(s) of chromosomes between a pair, selected arbitrarily from a population.
Machine Learning: Study on capability of learning without explicating programs.