nach oben

2011 | Buch

Kapitel lesen Erstes Kapitel lesen

Intelligent Systems

A Modern Approach

verfasst von: Crina Grosan, Ajith Abraham

Verlag: Springer Berlin Heidelberg

Buchreihe : Intelligent Systems Reference Library

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

Computational intelligence is a well-established paradigm, where new theories with a sound biological understanding have been evolving. The current experimental systems have many of the characteristics of biological computers (brains in other words) and are beginning to be built to perform a variety of tasks that are difficult or impossible to do with conventional computers. As evident, the ultimate achievement in this field would be to mimic or exceed human cognitive capabilities including reasoning, recognition, creativity, emotions, understanding, learning and so on. This book comprising of 17 chapters offers a step-by-step introduction (in a chronological order) to the various modern computational intelligence tools used in practical problem solving. Staring with different search techniques including informed and uninformed search, heuristic search, minmax, alpha-beta pruning methods, evolutionary algorithms and swarm intelligent techniques; the authors illustrate the design of knowledge-based systems and advanced expert systems, which incorporate uncertainty and fuzziness. Machine learning algorithms including decision trees and artificial neural networks are presented and finally the fundamentals of hybrid intelligent systems are also depicted.

Academics, scientists as well as engineers engaged in research, development and application of computational intelligence techniques, machine learning and data mining would find the comprehensive coverage of this book invaluable.

Inhaltsverzeichnis

Frontmatter

Evolution of Modern Computational Intelligence

Introduction

A conventional computational intelligence book introduction starts, many a times, with a history of Artificial Intelligence (AI) and what has been done up to date. This book introduction will try to start with what can be envisaged as future Computer Intelligence.

Crina Grosan, Ajith Abraham

Problem Solving by Search

Introduction

An important aspect of intelligence is goal-based problem solving. Several problems can be formulated as finding a sequence of actions that lead to a desirable goal. Each action changes the state and the aim is to find the sequence of actions and states that lead from the initial state to a final (goal) state.

Crina Grosan, Ajith Abraham

Informed (Heuristic) Search

Introduction

In the previous Chapter, we have presented several blind search or uninformed search techniques. Uninformed search methods systematically explore the search space until the goal is reached. As evident, uninformed search methods pursue options that many times lead away from the goal. Even for some small problems the search can take unacceptable amounts of time and/or space. The blind search techniques lack knowledge about the problem to solve and this makes them inefficient in many cases. Using problem specific knowledge can significantly improve the search speed.

Crina Grosan, Ajith Abraham

Iterative Search

Introduction

This chapter continues with the presentation of other informed search strategies (which are heuristics). They appear to be very useful for certain kind of problems even tough for certain categories of problems the quality of solution(s) provided may be unsatisfactory.

Crina Grosan, Ajith Abraham

Adversarial Search

Introduction

In the Chapters 2-4, we presented single agent search methods, that is, we only have one player, which has to move, without depending on the moves of another player (or players) and without competing or collaborating with any other players. This type of search is single agent search, and, naturally, multi-agent search is there in its turn.

Crina Grosan, Ajith Abraham

Knowledge Representation and Reasoning

Introduction

What is knowledge?

In a very generic way, knowledge can be defined as information (which can be expressed in the form of propositions) from the environment.

What is knowledge representation?

Again, in simple words, can be defined as symbols used to represent the propositions.

What is knowledge representation and reasoning?

One way to define it is as the manipulation of symbols encoding propositions to produce representations of new propositions.

The question of representing knowledge is a key issue in artificial intelligence: how can human knowledge of all kinds be represented by a computer language, and in such a way that computers can use this knowledge for purposes of reasoning? Modern computer applications have led to generalized use of knowledge representations in various contexts, including information search, simulation, web semantic ontology description.

Crina Grosan, Ajith Abraham

Rule-Based Expert Systems

Introduction

Rule-based systems (also known as production systems or expert systems) are the simplest form of artificial intelligence. A rule based system uses rules as the knowledge representation for knowledge coded into the system [1][3][4] [13][14][16][17][18][20]. The definitions of rule-based system depend almost entirely on expert systems, which are system that mimic the reasoning of human expert in solving a knowledge intensive problem. Instead of representing knowledge in a declarative, static way as a set of things which are true, rule-based system represent knowledge in terms of a set of rules that tells what to do or what to conclude in different situations.

Crina Grosan, Ajith Abraham

Managing Uncertainty in Rule Based Expert Systems

What Is Uncertainty and How to Deal With It?

Uncertainty is essentially lack of information to formulate a decision. The presence of uncertainty may result in making poor or bad decisions. In our daily life, as human beings, we are accustomed to dealing with uncertainty – that’s how we survive.

Crina Grosan, Ajith Abraham

Fuzzy Expert Systems

Introduction

One of the imprecision types of information encountered in an expert system is due to the (natural) language used to express information. If knowledge is not expressed in some formal language, the meaning cannot be interpreted exactly. Since there is no such universal scheme for formal representation language, a particular knowledge representation scheme must be chosen to adequately capture the information about the domain. Many a times the scheme chosen will not provide an exact match with the expert’s knowledge. Thus, imprecision will occur.

Crina Grosan, Ajith Abraham

Machine Learning

Introduction

Machine Learning[6][8][12] is concerned with the study of building computer programs that automatically improve and/or adapt their performance through experience. Machine learning can be thought of as “programming by example” [11]. Machine learning has many common things with other domains such as statistics and probability theory (understanding the phenomena that have generated the data), data mining (finding patterns in the data that are understandable by people) and cognitive sciences (human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people such as concept learning, skill acquisition, strategy change, etc.) [1].

Crina Grosan, Ajith Abraham

Decision Trees

Introduction

Decision trees are suitable for scientific problems entail labeling data items with one of a given, finite set of classes based on features of the data items. Decision Trees are classifiers that predict class labels for data items [3]. A decision tree learning algorithm approximates a target concept using a tree representation, where each internal node corresponds to an attribute, and every terminal node corresponds to a class[5][6][10].

Crina Grosan, Ajith Abraham

Artificial Neural Networks

Introduction

Artificial Neural Networks (ANN) are inspired by the way biological neural system works, such as the brain process information. The information processing system is composed of a large number of highly interconnected processing elements (neurons) working together to solve specific problems. ANNs, just like people, learn by example. Similar to learning in biological systems, ANN learning involves adjustments to the synaptic connections that exist between the neurons.

Crina Grosan, Ajith Abraham

Advanced Artificial Neural Networks

Introduction

The networks discussed in the previous Chapter – perceptron and multilayer perceptron – are feedforward networks in the sense that the information is processed forward from layer to layer and no cycles are presented in the network. Recurrent (or feedback) networks contain cycles which can connect neurons in the hidden layers with neurons in the input layers, hidden neurons between them or, in the most general case, all the neurons between them. A network may be fully recurrent, i.e., all units are connected back to each other and to themselves, or some part of the network may be fed back in recurrent links.

Crina Grosan, Ajith Abraham

Evolutionary Algorithms

Introduction

In nature, evolution is mostly determined by natural selection of different individuals competing for resources in the environment. Those individuals that are better are more likely to survive and propagate their genetic material. The encoding for genetic information (genome) is done in a way that admits asexual reproduction, which results in offspring that are genetically identical to the parent. Sexual reproduction allows some exchange and re-ordering of chromosomes, producing offspring that contain a combination of information from each parent. This is the recombination operation, which is often referred to as crossover because of the way strands of chromosomes cross over during the exchange. The diversity in the population is achieved by mutation operation.

Crina Grosan, Ajith Abraham

Evolutionary Metaheuristics

Introduction

Evolution Strategies (ES) were developed in [3][4]. ES tend to be used for empirical experiments that are difficult to model mathematically. The system to be optimized is actually constructed and ES are used to find the optimal parameter settings. Evolution strategies merely concentrate on translating the fundamental mechanisms of biological evolution for technical optimization problems. The parameters to be optimized are often represented by a vector of real numbers. Another vector of real numbers defines the strategy parameters, which controls the mutation of the objective parameters. Both object and strategic parameters form the data-structure for a single individual.

Crina Grosan, Ajith Abraham

Swarm Intelligence

Introduction

Swarm behavior can be seen in bird flocks, fish schools, as well as in insects like mosquitoes and midges. Many animal groups such as fish schools and bird flocks clearly display structural order, with the behavior of the organisms so integrated that even though they may change shape and direction, they appear to move as a single coherent entity [6].

Crina Grosan, Ajith Abraham

Hybrid Intelligent Systems

Introduction

Computational intelligence is an innovative framework for constructing intelligent hybrid architectures involving Neural Networks (NN), Fuzzy Inference Systems (FIS), Probabilistic Reasoning (PR), Evolutionary Computation (EC) and Swarm Intelligence (SI). Most of these hybridization approaches, however, follow an ad hoc design methodology, justified by success in certain application domains. Due to the lack of a common framework it often remains difficult to compare the various hybrid systems conceptually and to evaluate their performance comparatively.

Crina Grosan, Ajith Abraham

Titel: Intelligent Systems
verfasst von: Crina Grosan
Ajith Abraham
Verlag: Springer Berlin Heidelberg
Electronic ISBN: 978-3-642-21004-4
Print ISBN: 978-3-642-21003-7
DOI: https://doi.org/10.1007/978-3-642-21004-4