Michael W. Berry and Murray Browne

Understanding Search Engines: Mathematical Modeling and Text Retrieval

Software, Environments, and Tools 8

Understanding Search Engines discusses many of the key design issues for building search engines and emphasizes the important roles that applied mathematics can play in improving information retrieval. The authors discuss not only important data structures, algorithms, and software but also user-centered issues such as interfaces, manual indexing, and document preparation.

The authors bridge the gap between applied mathematics and information retrieval. They discuss some of the current problems in information
retrieval that may not be familiar to applied mathematicians and computer scientists and present some of the driving computational methods (SVD, SDD) for automated conceptual indexing.

This book uses a new approach to the subject by introducing topics in a nontechnical way and provides insights into common problems found in
information retrieval. The more mathematical details are provided in sidebars or are offset from the regular text.

Audience

Prerequisites include a year of undergraduate mathematics (calculus) and an introductory level course in linear algebra. The information science
aspects do not necessarily require formal coursework in information retrieval, but some internet knowledge and experience is necessary. The book
serves as an excellent companion text for courses in information retrieval, applied linear algebra, and scientific computing. Database managers
wishing to redesign a company's information retrieval system and scientists wanting to build intelligent indexing systems for large text collections
will find this volume essential.

About the Authors

Michael W. Berry is Associate Professor of Computer Science at the University of Tennessee. He is a member of SIAM, ACM, and the IEEE
Computer Society. He is coauthor of Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods (SIAM, 1993). Murray
Browne is a Research Associate in the Department of Computer Science at the University of Tennessee. He is a member of ASIS and has published
numerous essays, book reviews, and feature stories.

Contents

Chapter 1: Introduction; Chapter 2: Document File Preparation; Chapter 3: Vector Space Models; Chapter 4: Matrix Decompositions; Chapter 5:
Query Management; Chapter 6: Ranking and Relevance Feedback; Chapter 7: User Interface Considerations; Chapter 8: A Course Project; Chapter
9: Further Reading; Bibliography; Index.

1999 / xiv + 116 pages / Softcover / ISBN 0-89871-437-0


Michael E. Henderson, Christopher R. Anderson, and Stephen L. Lyons, editors

Object Oriented Methods for Interoperable Scientific and Engineering Computing

Proceedings in Applied Mathematics 99

There is a growing awareness in universities and industrial and government laboratories that object oriented methods have the potential to greatly improve the usefulness of computers in science and engineering. There are many efforts underway to redesign and re-implement large codes written in the 1970s and 1980s to take advantage of the improvement in maintainability and flexibility that object oriented designs offer.
Repositories such as Netlib and indices like GAMS have improved our ability to share code, but making the shared code useful requires widespread agreement about how the code is structured and how scientific and engineering codes should interoperate.

This volume contains papers presented at the October 1998 SIAM Workshop on Object Oriented Methods for Interoperable Scientific and
Engineering Computing that covered a variety of topics and issues related to designing and implementing computational tools for science and
engineering. Some examples include tools for ODEs and PDEs, discussions of how to write abstract code without loss of performance, and practical advice based on experiences.

The book includes experiences of both developers and industrial users of software, highlighting the difficult issues and merits of different approaches used in the aircraft, automotive, and petroleum industries, as well as national laboratories. There are also "real-world" papers in which authors have used spreadsheets, problem-solving environments like MATLAB, and other non-API interfaces to meet the demands of engineering user communities.

Audience

Anyone who is involved in developing or implementing mathematical algorithms in scientific and engineering computing will find this book valuable.

About the Editors

Michael E. Henderson is a research staff member in the Department of Mathematical Sciences at the IBM Thomas J. Watson Research Center.
Christopher R. Anderson is Professor of Mathematics at UCLA. Stephen L. Lyons is Senior Research Advisor for Mobil Technology Company.

Contents

Chapter 1: Current and Future Status of HPC in the World Automotive Industry, Myron Ginsberg; Chapter 2: General Exchange of Methods and
Data at Lockheed Martin, Michael R. Yokell and William P. Pfeiffer; Chapter 3: Exploiting Existing Software in Libraries: Successes, Failures, and
Reasons Why, William Gropp; Chapter 4: Language Interoperability Mechanisms for High-Performance Scientific Applications, Andrew Cleary, Scott
Kohn, Steven G. Smith, and Brent Smolinski; Chapter 5: Object Oriented Methods Without Object Oriented Languages: Can Intermediate
Approaches Facilitate the Adoption of OO Methods in the Research Community?, David E. Bernholdt; Chapter 6: Developing an Integrated
Environment for Computational Field Simulation, Samuel H. Russ and Adam Gaither; Chapter 7: A Microkernel Design for Component-based Parallel
Numerical Software Systems, Satish Balay, Bill Gropp, Lois Curfman McInnes, and Barry Smith; Chapter 8: When the "One Size Fits All'' Doesn't
Fit, Michael E. Henderson; Chapter 9: Flexibility and Interoperability in a Parallel Biomolecular Dynamics Code, Robert Brunner, Laxmikant Kal・
and James Phillips; Chapter 10: An Object-Oriented Approach for Development and Testing of Parallel Solution Algorithms for Nonlinear PDEs,
Richard D. Hornung and Carol S. Woodward; Chapter 11: Development and Utilization of Parallel Generic Algorithms for Scientific Computations,
Atanas Radenski, Andrew Vann, and Boyana Norris; Chapter 12: Design of the hypre Preconditioner Library, Edmond Chow, Andrew J. Cleary, and
Robert D. Falgout; Chapter 13: Generic Programming for High Performance Numerical Linear Algebra, Jeremy G. Siek, Andrew Lumsdaine,
Lie-Quan Lee; Chapter 14: Developing a Derivative-Enhanced Object-Oriented Toolkit for Scientific Computations, Paul Hovland, Boyana Norris,
Lucas Roh, and Barry Smith; Chapter 15: Algorithm Development for Large Scale Computing, Matthew G. Knepley and Vivek Sarin; Chapter 16:
Evolution of the NAG Library ODE Solvers, I. Gladwell; Chapter 17: Object Oriented Toolbox for Modelling and Simulation of Dynamical Systems,
Mikael Zebbelin Poulsen, Falko Jens Wagner, Per Grove Thomsen, and Neils Houbek; Chapter 18: Design and Implementation of an Object Oriented
C++ Library for Nonlinear Optimization, David L. Bruhwiler, Svetlana G. Shasharina, John R. Cary, and David Alexander; Chapter 19: ADMAT:
Automatic Differentiation in MATLAB Using Object Oriented Methods, Arun Verma; Chapter 20: Object-Oriented Programming for General Mixed
Finite Element Methods, Tong Sun, Richard E. Ewing, Hongsen Chen, Stephen L. Lyons, Guan Qin; Chapter 21: The Design of a Finite
Element/Spectral Element Code for Incompressible Fluid Flow, Einar M. Ronquist; Chapter 22: Programming Engineering Applications Using the
Object Oriented FEM Code Castem 2000, Andrei Constantinescu, Marta Dragon, and Joel Kichenin; Chapter 23: Mesh Component Design and
Software Integration Within SUMAA3d, Lori Frietag, Mark Jones, and Paul Plassmann; Chapter 24: PDESolve: An Object-Oriented PDE Analysis
Environment, Kevin Long and Brian Van Straalen; Chapter 25: The Use of Object-Oriented Design Patterns in the SAMRAI Structured AMR
Framework, Richard D. Hornung and Scott R. Kohn; Chapter 26: Overture: An Object-Oriented Framework for Solving Partial Differential Equations
on Overlapping Grids, David L. Brown, William D. Henshaw, and Daniel J. Quinlan; Chapter 27: VBM and MCCC: Packages for Object Oriented
Visualization and Computation of Bifurcation Manifolds, Randy C. Paffenroth; Chapter 28: Experiences Developing ALEGRA: A C++ Coupled
Physics Framework, Kent G. Budge and James S. Peery; Chapter 29: Rapid Application Development and Enhanced Code Interoperability using the
POOMA Framework, Julian C. Cummings, James A. Crotinger, Scott W. Haney, William F. Humphrey, Steve R. Karmesin, John V.W. Reynders,
Stephen A. Smith, and Timothy J. Williams; Chapter 30: Active Libraries: Re-thinking the Roles of Compilers and Libraries, Todd L. Veldhuizen and
Dennis Gannon; Chapter 31: Application Oriented Library Design, Geoffrey M. Furnish; Chapter 32: Optimizations for Parallel Object-Oriented
Frameworks, Fede Bassetti, Kei Davis, and Daniel Quinlan; Chapter 33: Distributed Computing: What Do We Need and Can We Get It with Java?,
Christopher R. Anderson.

1999 / xiv + 321 pages / Softcover / ISBN 0-89871-445-1


by Yuen Chung Kwong (National University of Singapore)

ANNUAL REVIEW OF SCALABLE COMPUTING

This book provides a forum for researchers in scalable computing to publish extended-length articles on significant new developments. An article may present comprehensive results from a major project, review recent work in a sub-domain, or expound new ideas in a detailed, tutorial fashion, at a length which most journals and conference proceedings cannot accommodate.

The five articles in this book give an excellent illustration of the different types of material requiring such extensive treatment, and should serve well to encourage future authors with similar ideas to consider publishing in the Series on Scalable Computing.

Contents:

Active Objects: A Software Structure for Cluster-Based Systems (C K Yuen)
Scalable Optimistic Parallel Simulation (Y M Teo & S C Tay)
High Performance Fortran for Advanced Applications (S Benkner)
Interprocess Communication Optimization in a Scalable Computing Cluster (O La'adan & A Barak)
Designing Superservers with Clusters and Commodity Components (Z Xu & K Hwang)

Readership: Researchers in computer science.

210pp (approx.)
Pub. date: Scheduled Autumn 1999
ISBN 981-02-4119-4


by Larry Wos & Gail W Pieper (Argonne National Laboratory)

A FASCINATING COUNTRY IN THE WORLD OF COMPUTING
Your Guide to Automated Reasoning

This book shows you ・through examples and puzzles and intriguing questions ・how to make your computer reason logically. To help you, the book includes a CD-ROM with OTTER, the world's most powerful general-purpose reasoning program. The automation of reasoning has advanced markedly in the past few decades, and this book discusses some of the remarkable successes that automated reasoning programs have had in tackling challenging problems in mathematics, logic, program verification, and circuit design. Because the intended audience includes students and teachers, the book provides many exercises (with hints and also answers), as well as tutorial chapters that gently introduce readers to the field of logic and to automated reasoning in general. For more advanced researchers, the book presents challenging questions, many of which are still unsolved.

Contents:

Introduction and Map for Reading the Book
Learning Logic by Example
A Brisk Introduction to Automated Reasoning
Logic Circuit Design
Logic Circuit Verification
Research in Mathematics and Logic
Formal Underpinnings
Guidelines for OTTER
Vignettes Focusing on the Companion Book, The Collected Works of Larry Wos
Open Questions
Appendixes with Input and Output Files and Proofs

Readership: College students, teachers, researchers and historians of computer science.

608pp (approx.)
Pub. date: Scheduled Autumn 1999
ISBN 981-02-3910-6


Advanced Series in Mathematical Physics

INVERSE SCATTERING FOR WAVE AND DIFFUSION EQUATIONS
(with CD-ROM)

by J W Wiskin, D T Borup, S A Johnson (University of Utah) & D N Ghosh Roy (Sachs & Freeman Inc.)

The books presently available on Inverse Scattering are written primarily for those students who have a fairly substantial mathematical background. This book is directed towards students in the BioEngineering and Environmental Engineering sciences who do not have this optimal exposure to mathematical concepts.

This book:
Includes practical algorithms on CD-ROM for forward simulation and inversion of wave propagation in 2D.

Can be used in the classroom to teach the concepts of inverse scattering and imaging, and yet brings the student to the forefront of research in this important area.

Discusses in clear terms the mathematical preliminaries that are required for the proper understanding of the inversion/imaging problem, as they are required.

Couples the theoretical development with a practical discussion of data completeness requirements and special calibration methods for EM and acoustic scattering to
get reasonable images from laboratory data.

Includes careful discussion of the case of acoustic wave propagation in biological tissue (human breast for example). After this introduction, the more difficult cases ofElectromagnetic waves in non-magnetic media are considered. These require discussion of the vector wave equation, and the numerical methods used in their solution.The methods chosen are the Finite Difference Time Domain Method, the Frequency Domain Integral Equation method, and the associated Hybrid method, whichcombines the particular advantages of both of these methods.

Discusses the important case of the diffusion (quasi-static) approximation to the EM wave equations, and its many applications. Such applications include: 1) location ofburied mines in surf zones, 2) location and delineation of hazardous waste leakage in waste disposal sites (Dense Non-Aqueous Phase Liquids ・DNAPL's 3) breastcancer imaging and 4) micro-impedance imaging of cells.

Readership: Students in biomedical, environmental, electrical and civil engineering, as well as applied physics.

300pp (approx.)
Pub. date: Scheduled Summer 2000
ISBN 981-02-4172-0
ISBN 981-02-4173-9(pbk)


Series on Applied Mathematics

INVERSE PROBLEMS FOR ELECTRICAL NETWORKS

by E B Curtis & J A Morrow (University of Washington, Seattle)

This book is a very timely exposition of part of an important subject which goes under the general name of "inverse problems". The analogous problem for continuous media has been very much studied, with a great deal of difficult mathematics involved, especially partial differential equations. Some of the researchers working on the inverse conductivity problem for continuous media (the problem of recovering the conductivity inside from measurements on the outside) have taken an interest in the authors' analysis of this similar problem for resistor networks.

The authors' treatment of inverse problems for electrical networks is at a fairly elementary level. It is accessible to advanced undergraduates, and mathematics students at the graduate level. The topics are of interest to mathematicians working on inverse problems, and possibly to electrical engineers. A few techniques from other areas of mathematics have been brought together in the treatment. It is this amalgamation of such topics as graph theory, medial graphs and matrix algebra, as well as the analogy to inverse problems for partial differential equations, that makes the book both original and interesting.

Contents:

Circular Planar Graphs
Resistor Networks
Harmonic Functions
Characterization I
Adjoining Edges
Characterization II
Medial Graphs
Recovering a Graph
Layered Networks

Readership: Graduate students and researchers in applied mathematics and electrical and electronic engineering.

200pp (approx.)
Pub. date: Scheduled Spring 2000
ISBN 981-02-4174-7


by Jan Lopuszanski (University of Wroclaw, Poland)

THE INVERSE VARIATIONAL PROBLEM IN CLASSICAL MECHANICS

This book provides a concise description of the current status of a fascinating scientific problem ・the inverse variational problem in classical
mechanics. The essence of this problem is as follows: one is given a set of equations of motion describing a certain classical mechanical system, and the question to be answered is: Do these equations of motion correspond to some Lagrange function as its Euler豊agrange equations? In general, not for every system of equations of motion does a Lagrange function exist; it can, however, happen that one may modify the given equations of motion in such a way that they yield the same set of solutions as the original ones and they correspond already to a Lagrange function. Moreover, there can even be infinitely many such Lagrange functions, the relations among which are not trivial. The book deals with this scope of problems. No advanced mathematical methods, such as, contemporary differential geometry, are used. The intention is to meet the standard educational level of a broad group of physicists and mathematicians. The book is well suited for use as lecture notes in a university course for physicists.

Contents:

Constants of Motion
Theorem of Henneaux
Instructive Example of Douglas
Construction of the Most General Autonomous One-Particle Lagrange Function in (3+1) Space-time Dimensions Giving Rise to Rotationally Covariant Euler-
Lagrange Equations
Evaluation of the Function Gij
Construction of the Most General Two-Particle Lagrange Function in (1+1) Space-time Dimensions Giving Rise to Euler-Lagrange Equations Covariant Under
Galilei Transformation
Galilei Forminvariance of the Euler-Lagrange Equations for Two Particles in (1+1) Space-time Dimensions

Readership: Graduate students of theoretical physics and mathematics, as well as theoretical physicists doing research in classical and quantum mechanics.

236pp
Pub. date: Nov 1999
ISBN 981-02-4178-X


by Lars Skyttner (University of G舸le, Sweden)

GENERAL SYSTEMS THEORY
Ideas and Applications

The world in which classical positivistic science and technology obtained great success has vanished. However, the way of thinking promoted by that epoch still lingers in our social consciousness, sometimes as a burden. To conquer the shortcomings of classical analytical science in the modern, ever more complex world, systems theory and its applications within systems science present an alternative to old paradigms.

Systems theorists see common principles in the structure and operation of systems of all kinds and sizes. They promote an interdisciplinary science adapted for a universal application with a common language and area of concepts. This approach is seen as a means of not only overcoming the fragmentation of knowledge and the isolation of the specialist, but also finding new solutions to problems created by the earlier "solution of problems".

This book introduces the systemic alternative. It is divided into two parts. The first is devoted to the historical background of the systems movement, and presents pioneering thoughts and theories of the area. Basic concepts of general systems theory with well-known laws and principles are discussed, as well as related topics like cybernetics and information theory.

The second part deals with some of the common applications of systems theory within systems science, such as artificial intelligence, management information systems and informatics. An attempt is made to predict the future of systems theory in a world apparently becoming fragmented and integrated at the same time.

To engage oneself in systems theory and its striving towards an applied universal science is a highly cross-scientific occupation. The reader will come into contact with many different academic disciplines, and consequently the possibility of an all-round education ・something particularly needed in our over-specialized world.

Contents:

The Ideas and Why: The Emergence of Holistic Thinking
Basic Ideas of Systems
A Selection of Systems Theories
Information and Communication Theory
Some Theories of Brain and Mind
The Applications and How: Artificial Intelligence and Life
Organization Theory and Management Cybernetics
Decision Making and Decision Aids
Informatics
Some of the Systems Methodologies
The Future of Systems Theory

Readership: Computer specialists, architects, businessmen, teachers and holistic thinkers.

330pp (approx.)
Pub. date: Scheduled Spring 2000
ISBN 981-02-4175-5
ISBN 981-02-4176-3(pbk)