An Intelligent Process Development Using Fusion of Genetic Algorithm with Fuzzy Logic

An Intelligent Process Development Using Fusion of Genetic Algorithm with Fuzzy Logic

Kunjal Bharatkumar Mankad (Independent Researcher, India)
DOI: 10.4018/978-1-4666-7258-1.ch002
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Intelligent System (IS) can be defined as the system that incorporates intelligence into applications being handled by machines. The chapter extensively discusses the role of Genetic Algorithm (GA) in the search and optimization process along with discussing applications developed so far. A very detailed discussion on the Fuzzy Rule-Based System is presented along with major applications developed in different domains. The chapter presents algorithm of implementing intelligent procedure to decide whether a patient is prone to heart disease or not. The procedure evolves solutions using genetic operators and provides its decision automatically. The chapter presents discussion on the results achieved as a result of prototypical implementation of the evolutionary fuzzy hybrid model. The significant advantage of the presented research work is that applications that do not have any mathematical formulation and still demand optimization can be easily solved using the designed approach.
Chapter Preview
Top

Introduction

Current scenario of real life applications development is a result of utilizing intelligent techniques provided by soft computing family. The research work focuses on evolutionary computing for searching the optimum solution from global search space. Evolutionary computing provides the way of achieving optimized outcome with less complexity and high accuracy. Due to the above stated qualities, Evolutionary Computing is utilized as major computing approach in search and optimization. Evolutionary Computing has major four constituents such as Genetic Algorithm (GA), Genetic Programming (GP), Evolutionary Programming (EP) and Evolutionary Strategies (ES). Genetic Algorithm is prime constituent of Evolutionary Computing (EC). GA provides efficient search and optimized outcome. But at the same time GA suffers from two major limitations such as lack of domain knowledge representation and handling imprecision and uncertainty. In order to deal with the stated limitations of GA, Fuzzy Logic is hybridized with GA. The chapter proposes machine learning approach using genetic-fuzzy hybridization for medical diagnostic application for heart disease identification. The current research work provides a novel design of genetic-fuzzy hybrid structure for automatic decision support. The chapter presents the design of intelligent procedure that supports genetics based rule learning which integrates linguistic knowledge representation. The chapter presents algorithm of implementing intelligent procedure along with discussion of every steps. The chapter explains different parameters of patent’s cardiac profile in order to decide whether patient is prone to heart disease or not. Supervised machine learning is implemented to incorporate medical expert’s knowledge into the system. Machine learning is made possible by utilizing the developed intelligent process using Genetic Algorithm and Fuzzy Logic. The research work is based on supervised machine learning approach in which machine is trained through expert’s knowledge once, and later on takes decision automatically every time. The outcome provided by Genetic Algorithm is mapped to get automatic decision for real life application. It presents discussion on the results achieved as prototypical implementation of evolutionary fuzzy hybrid model for medical diagnostic application. The significant advantage of the presented research work is that applications which do not have any mathematical formulation and still demands optimization can be easily solved using the designed approach.

Key Terms in this Chapter

Intelligent System: Intelligent System (IS) can be defined as the system that incorporates intelligence into applications being handled by machines. Intelligent systems perform search and optimization along with learning capabilities. Different types of machine learning such as supervised, unsupervised and reinforcement learning can be modeled in designing intelligent systems. Intelligent systems also perform complex automated tasks which are not possible by traditional computing paradigm. Various diagnostic, robotics and engineering systems are results of intelligent procedures implemented in Intelligent System Design.

Genetic Operators: The mathematical formula applied as a step of genetic algorithm is known as operator. Selection, Crossover and Mutation operators are basic genetic operators. They are the mechanism to generate evolution from one generation to another generation. These operators are application specific. Crossover is of prime type of genetic operators and available in many types such as single point crossover, two point crossover and multipoint crossover. Machine learning methods and optimization methods are utilizing them in order to achieve automatic evolution for intelligent system design.

Genetic Algorithms: Algorithms based on principles of Darwinian evolution (natural evolution). They are successfully applied to the problems which are difficult to solve using conventional techniques. Machine learning and optimization effectively use Genetic Algorithms. They are basically search algorithms but can be applied to model learning tasks also. Robotics and swarm intelligent systems as well as evolutionary systems are taking advantages of parallel processing and multi-objective optimization due to characteristics of Genetic Algorithms. Genetic Algorithms are widely used in engineering, scientific and business applications.

Fuzzy System: A fuzzy system (FS) is any Fuzzy Logic based system, which either uses Fuzzy Logic as the basis for knowledge representation using different forms of knowledge. Systems variables interactions and intermodal relationship can be modeled using different ways such as membership functions and shape analysis of membership function. Membership functions are mathematical way of representing values for inference mechanism.

Evolutionary Algorithms: Evolutionary Algorithms are inspired by natural evolution. They are especially associated to Artificial Intelligence (AI) search techniques. EAs are computer programs that attempt to solve complex problems by mimicking the processes of Darwinian evolution. The evolutionary system performs evolutionary computing. Evolutionary computing (EC) is the basis of Evolutionary Algorithms (EA).

Soft Computing: A set of artificial intelligence techniques provides efficient and feasible solutions in comparison with conventional computing. These techniques are also known as computational intelligence. They are basically integrated techniques to find solutions for the problems which are highly complex, ill- defined and difficult to model. Real world problems deal with imprecision and uncertainty can be easily handled using such techniques. Soft computing provides set of techniques which are hybridized and finally useful for designing intelligent systems.

Evolutionary Computing: Evolutionary Computing is the study of computational systems which use ideas and get inspirations from natural evolution and other biological systems. These types of computing techniques are basically designed for evolution of characteristics inherited from one generation to another generation. They are computer based problem solving systems which incorporate computational models of evolutionary processes. EC provides four main methods namely Genetic Algorithms (GA), Evolutionary Strategies (ES), Evolutionary Programming (EP) and Genetic Programming (GP).

Hard Computing: Traditional computing techniques based on principles of precision, uncertainty and rigor. The problems based on analytical model can be easily solved using such techniques. Real world problems which deal with changing of information and imprecise behavior can- not be handled by hard computing techniques.

Complete Chapter List

Search this Book:
Reset