Foundations of Computational Intelligence Volume 2: Approximate Reasoning (Studies in Computational Intelligence (202))

Annotation Human reasoning usually is very approximate and involves various types of uncertainties. Approximate reasoning is the computational modelling of any part of the process used by humans to reason about natural phenomena or to solve real world problems. The scope of this book includes fuzzy sets, Dempster-Shafer theory, multi-valued logic, probability, random sets, and rough set, near set and hybrid intelligent systems. Besides research articles and expository papers on theory and algorithms of approximation reasoning, papers on numerical experiments and real world applications were also encouraged. This Volume comprises of 12 chapters including an overview chapter providing an up-to-date and state-of-the research on the applications of Computational Intelligence techniques for approximation reasoning. The Volume is divided into 2 parts: Part-I: Approximate Reasoning Theoretical Foundations and Part-II: Approximate Reasoning Success Stories and Real World Applications Annotation The decision to invest in oil field development is an extremely complex problem, even in the absence of uncertainty, due to the great number of technological alternatives that may be used, to the dynamic complexity of oil reservoirs - which involves multiphase flows (oil, gas and water) in porous media with phase change, and to the complicated combinatorial optimization problem of choosing the optimal oil well network, that is, choosing the number and types of wells (horizontal, vertical, directional, multilateral) required for draining oil from a field with a view to maximizing its economic value. The present book is a result of about 4 years of research in this area through a partnership between the Applied Computational Intelligence Laboratory (ICA) of the Department of Electrical Engineering at PUC-Rio, and Petrobras, through its R & D (research and development) program called PRAVAP (Advanced Oil Recovery Program), which is linked to its research center (CENPES). The book makes use of computational intelligence techniques, especially genetic algorithms, genetic programming, neural networks, fuzzy logic and neuro-fuzzy systems for purposes of solving this investment under uncertainty problem. These techniques are combined with modern finance theory, particularly with the real options theory, also known as the investment under uncertainty theory, in such a way as to provide practical as well as theoretically rigorous solutions. This partnership, through which countless master's and doctoral theses were produced at PUC-Rio and computational methodologies and programs were developed for Petrobras, has been summarized in this original and comprehensive work, now available to a wider audience of researchers and interested readers Foundations of Computational Intelligence Volume 2: Approximation Reasoning: Theoretical Foundations and Applications Human reasoning usually is very approximate and involves various types of - certainties. Approximate reasoning is the computational modelling of any part of the process used by humans to reason about natural phenomena or to solve real world problems. The scope of this book includes fuzzy sets, Dempster-Shafer theory, multi-valued logic, probability, random sets, and rough set, near set and hybrid intelligent systems. Besides research articles and expository papers on t- ory and algorithms of approximation reasoning, papers on numerical experiments and real world applications were also encouraged. This Volume comprises of 12 chapters including an overview chapter providing an up-to-date and state-of-the research on the applications of Computational Intelligence techniques for - proximation reasoning. The Volume is divided into 2 parts: Part-I: Approximate Reasoning – Theoretical Foundations Part-II: Approximate Reasoning – Success Stories and Real World Applications Part I on Approximate Reasoning – Theoretical Foundations contains four ch- ters that describe several approaches of fuzzy and Para consistent annotated logic approximation reasoning. In Chapter 1, “Fuzzy Sets, Near Sets, and Rough Sets for Your Computational Intelligence Toolbox” by Peters considers how a user might utilize fuzzy sets, near sets, and rough sets, taken separately or taken together in hybridizations as part of a computational intelligence toolbox. In multi-criteria decision making, it is necessary to aggregate (combine) utility values corresponding to several criteria (parameters). Chaos is a fascinating phenomenon that has been observed in nature, laboratory, and has been applied in various real-world applications. Chaotic systems are deterministic with no random elements involved yet their behavior appears to be random. Observations of chaotic behavior in nature include weather and climate, the dynamics of satellites in the solar system, the time evolution of the magnetic field of celestial bodies, population growth in ecology, to mention only a few examples. Chaos has been observed in the laboratory in a number of systems such as electrical circuits, lasers, chemical reactions, fluid dynamics, mechanical systems, and magneto-mechanical devices. Chaotic behavior has also found numerous applications in electrical and communication engineering, information and communication technologies, biology and medicine. The book project was launched to access the latest research related to chaos applications in intelligent computing where researchers from all over the world provide the necessary coverage of the mentioned field. The primary objective of this project was to assemble as much research coverage as possible related to the field by defining the latest innovative technologies and providing the most comprehensive list of research references. The coverage of this book provides strength to this reference resource for both researchers and also decision makers in obtaining a greater understanding of the concepts, issues, problems, trends, challenges and opportunities related to this field of study Information granules and their processing permeate a way in which we perceive the world, carryout processing at the conceptual (abstract) level, and communicate our findings to the surrounding environment. The importance of information granulation becomes even more apparent when we are faced with a rapidly growing flood of data, become challenged to make decisions in complex data settings and are required to appreciate the context from which the data is derived. Human centricity of systems that claim to be “intelligent” and the granular computing come hand in hand. It is not surprising at all to witness that the paradigm of Granular Computing has started to gain visibility and continues along this path by gathering interest from the circles of academics and practitioners. It is quite remarkable that the spectrum of application and research areas that have adopted information granulation as a successful strategy for dealing with information complexity covers such diverse fields as bioinformatics, image understanding, environmental monitoring, urban sustainability, to mention few most visible in the literature. Undoubtedly, there are two important aspects of Granular Computing that are worth stressing. First, there are several formalisms in which information granules are articulated so be intervals (sets), fuzzy sets, rough sets, soft sets, approximate sets, near sets and alike. They are complementary and each of them offers some interesting views at the complexity of the world and cyberspace. The decision to invest in oil field development is an extremely complex problem, even in the absence of uncertainty, due to the great number of technological alternatives that may be used, to the dynamic complexity of oil reservoirs - which involves mul- phase flows (oil, gas and water) in porous media with phase change, and to the c- plicated combinatorial optimization problem of choosing the optimal oil well network, that is, choosing the number and types of wells (horizontal, vertical, directional, m- tilateral) required for draining oil from a field with a view to maximizing its economic value. This problem becomes even more difficult when technical uncertainty and e- nomic uncertainty are considered. The former are uncertainties regarding the existence, volume and quality of a reservoir and may encourage an investment in information before the field is developed, in order to reduce these uncertainties and thus optimize the heavy investments required for developing the reservoir. The economic or market uncertainties are associated with the general movements of the economy, such as oil prices, gas demand, exchange rates, etc., and may lead decision-makers to defer - vestments and wait for better market conditions. Choosing the optimal investment moment under uncertainty is a complex problem which traditionally involves dynamic programming tools and other techniques that are used by the real options theory. Scene modeling is a very important part in Computer Graphics because it allows c- ating more or less complex models to be rendered, coming from the real world or from the designer's imagination. However, scene modeling is a very difficult task, as there is a need of more and more complex scenes and traditional geometric modelers are not well adapted to computer aided design. Even if traditional scene modelers offer very interesting tools to facilitate the designer's work, they suffer from a very important drawback, the lack of flexibility, which does not authorize the designer to use incomplete or imprecise descriptions, in order to express his (her) mental image of the scene to be designed. Thus, with most of the current geometric modelers the user must have a quite precise idea of the scene to design before using the modeler to achieve the modeling task. This kind of design is not really a computer aided one, because the main creative ideas have been elaborated without any help of the modeler. Declarative scene modeling could be an interesting alternative to traditional g- metric modeling. Indeed, declarative scene modeling tries to give intuitive solutions to the scene modeling problem by using Artificial Intelligence techniques which allow the user to describe high level properties of a scene and the modeler to give all the solutions corresponding to imprecise properties. Chaos is a fascinating phenomenon that has been observed in nature, laboratory, and has been applied in various real-world applications. Chaotic systems are deterministic with no random elements involved yet their behavior appears to be random. Obser- tions of chaotic behavior in nature include weather and climate, the dynamics of sat- lites in the solar system, the time evolution of the magnetic field of celestial bodies, population growth in ecology, to mention only a few examples. Chaos has been observed in the laboratory in a number of systems such as electrical circuits, lasers, chemical reactions, fluid dynamics, mechanical systems, and magneto-mechanical devices. Chaotic behavior has also found numerous applications in electrical and communication engineering, information and communication technologies, biology and medicine. To the best of our knowledge, this is the first book edited on chaos applications in intelligent computing. To access the latest research related to chaos applications in intelligent computing, we launched the book project where researchers from all over the world provide the necessary coverage of the mentioned field. The primary obj- tive of this project was to assemble as much research coverage as possible related to the field by defining the latest innovative technologies and providing the most c- prehensive list of research references. "Intelligent Systems use a range of methodologies for analysis, pre-processing, storage, organization, enhancing and mining of operational data, turning it into useful information and knowledge for decision makers in business enterprises. These intelligent technologies for decision support have been used with success by companies and organizations that are looking for competitive advantages whenever the issues on forecast, optimization, risks analysis, fraud detection, and decision under uncertainties are presented." "Intelligent Systems (IS) offer to managers and decision makers the best solutions for complex applications, normally considered difficult, very restrictive or even impossible. The use of such techniques leads to a revolutionary process which has a significant impact in the business management strategy, by providing on time, correct information, ready to use. Computational intelligence techniques, especially genetic algorithms, genetic programming, neural networks, fuzzy logic and neuro-fuzzy as well as modern finance theories, such as real options theory, are here presented and exemplified in oil and gas exploitation and production. This book is addressed to executives and students, directly involved or interested in intelligent management in different fields." --Book Jacket Annotation The idea of Human-Centric Information Processing was prompted by the pioneering work of Zadeh Towards a theory of fuzzy information granulation and its centrality in human reasoning and fuzzy logic (1997). Since the publication of this work, a large number of researchers have focused on identifying the nature of information granulation and its specific relationship to human information processing. We now begin to witness the rich and manifold results of this concentrated research effort of the last decade. This volume is intended to document the milestone contributions to human-centric information processing research and to demonstrate the emerging computational methods and the processing environments that arose from these research insights. The chapters, written by experts in the field, cover the fundamental methodologies, the new information processing paradigm, functional architectures of granular information processing and granular modeling applications. The book provides a valuable reference for researchers, graduate students intending to focus on particular aspects of human-centric information processing from a broadly informed perspective and for practitioners to whom the breadth of coverage of topics will help inspire innovative applications Front Matter....Pages - Front Matter....Pages 1-1 Fuzzy Sets, Near Sets, and Rough Sets for Your Computational Intelligence Toolbox....Pages 3-25 Fuzzy without Fuzzy: Why Fuzzy-Related Aggregation Techniques Are Often Better Even in Situations without True Fuzziness....Pages 27-51 Intermediate Degrees Are Needed for the World to Be Cognizable: Towards a New Justification for Fuzzy Logic Ideas....Pages 53-74 Paraconsistent Annotated Logic Program Before-after EVALPSN and Its Application....Pages 75-108 Front Matter....Pages 109-109 A Fuzzy Set Approach to Software Reliability Modeling....Pages 111-131 Computational Methods for Investment Portfolio: The Use of Fuzzy Measures and Constraint Programming for Risk Management....Pages 133-173 A Bayesian Solution to the Modifiable Areal Unit Problem....Pages 175-196 Fuzzy Logic Control in Communication Networks....Pages 197-236 Adaptation in Classification Systems....Pages 237-258 Music Instrument Estimation in Polyphonic Sound Based on Short-Term Spectrum Match....Pages 259-273 Ultrasound Biomicroscopy Glaucoma Images Analysis Based on Rough Set and Pulse Coupled Neural Network....Pages 275-293 An Overview of Fuzzy C-Means Based Image Clustering Algorithms....Pages 295-310 Back Matter....Pages -