Elements of computational statistics

Sample

Vladimir S. Lerner

The Variation Principle In Informational Macrodynamics (VP) introduces the process' integral information measure, which has a distinctive difference from traditional information approaches that use an entropy function. The VP's minimax mathematical mechanisms formalize the regularities of the cooperative informational dynamics, connecting randomness and regularities, stochastic and determinism, reversibility and irreversibility, symmetry and nonsymmetry, stability and instability, regular and chaotic dynamics, thermodynamics and informational dynamics, time-reversible and time-irreversible processes, reveals the information dynamic mechanisms of evolution and development. The mathematical formalisms and methods, implemented in the forms of computer models, algorithms and programs, provide new general tools for Information Systems' Modeling in such areas as biology, intelligent systems, computer and information technology, including communications, data modeling and data management. Variation Principle In Informational Macrodynamics, with its examples and applications, will meet the needs of a professional audience composed of researchers and practitioners in industry, as well as graduate-level students in computer science, mathematics and engineering.

Sample

Sven Jörges

Automatic code generation is an essential cornerstone of model-driven approaches to software development. Currently, lots of techniques are available that support the specification and implementation of code generators, such as engines based on templates or rule-based transformations. All those techniques have in common that code generators are either directly programmed or described by means of textual specifications. This monograph presents Genesys, a general approach, which advocates the graphical development of code generators for arbitrary source and target languages, on the basis of models and services. In particular, it is designed to support incremental language development on arbitrary metalevels. The use of models allows building code generators in a truly platform-independent and domain-specific way. Furthermore, models are amenable to formal verification methods such as model checking, which increase the reliability and robustness of the code generators. Services enable the reuse and integration of existing code generation frameworks and tools regardless of their complexity, and at the same time manifest as easy-to-use building blocks which facilitate agile development through quick interchangeability. Both, models and services, are reusable and thus form a growing repository for the fast creation and evolution of code generators.

Sample

Walter Oleksy

Examines how computer technology affects the acquisition and application of scientific knowledge in such fields as astronomy, archaeology, medicine, and physics.

Sample

Witold Pedrycz

Granular Computing is concerned with constructing and processing carried out at the level of information granules. Using information granules, we comprehend the world and interact with it, no matter which intelligent endeavor this may involve. The landscape of granular computing is immensely rich and involves set theory (interval mathematics), fuzzy sets, rough sets, random sets linked together in a highly synergetic environment. This volume is a first comprehensive treatment of this emerging paradigm and embraces its fundamentals, underlying methodological framework, and a sound algorithmic environment. The panoply of applications covered includes system identification, telecommunications, linguistics and music processing. Written by experts in the field, this volume will appeal to all developing intelligent systems, either working at the methodological level or interested in detailed system realization.