Search

This volume provides approaches and solutions to challenges occurring at the interface of research fields such as, e.g., data analysis, data mining and knowledge discovery, computer science, operations research, and statistics. In addition to theory-oriented contributions various application areas are included. Moreover, traditional classification research directions concerning network data, graphs, and social relationships as well as statistical musicology describe examples for current interest fields tackled by the authors. The book comprises a total of 55 selected papers presented at the Joint Conference of the German Classification Society (GfKl), the German Association for Pattern Recognition (DAGM), and the Symposium of the International Federation of Classification Societies (IFCS) in 2011.​

This book constitutes the refereed proceedings of the 13th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2012, held in Natal, Brazil, in August 2012. The 100 revised full papers presented were carefully reviewed and selected from more than 200 submissions for inclusion in the book and present the latest theoretical advances and real-world applications in computational intelligence.

This book comprehensively assesses the growing importance of project data for project scheduling, risk analysis and control. It discusses the relevance of project data for both researchers and professionals, and illustrates why the collection, processing and use of such data is not as straightforward as most people think. The theme of this book is known in the literature as data-driven project management and includes the discussion of using computer algorithms, human intuition, and project data for managing projects under risk. The book reviews the basic components of data-driven project management by summarizing the current state-of-the-art methodologies, including the latest computer and machine learning algorithms and statistical methodologies, for project risk and control. It highlights the importance of artificial project data for academics, and describes the specific requirements such data must meet. In turn, the book discusses a wide variety of statistical methods available to generate these artificial data and shows how they have helped researchers to develop algorithms and tools to improve decision-making in project management. Moreover, it examines the relevance of project data from a professional standpoint and describes how professionals should collect empirical project data for better decision-making. Finally, the book introduces a new approach to data collection, generation, and analysis for creating project databases, making it relevant for academic researchers and professional project managers alike.

Implement numerical algorithms in Java using NM Dev, an object-oriented and high-performance programming library for mathematics.You'll see how it can help you easily create a solution for your complex engineering problem by quickly putting together classes.Numerical Methods Using Java covers a wide range of topics, including chapters on linear algebra, root finding, curve fitting, differentiation and integration, solving differential equations, random numbers and simulation, a whole suite of unconstrained and constrained optimization algorithms, statistics, regression and time series analysis. The mathematical concepts behind the algorithms are clearly explained, with plenty of code examples and illustrations to help even beginners get started. What You Will Learn Program in Java using a high-performance numerical libraryLearn the mathematics for a wide range of numerical computing algorithmsConvert ideas and equations into codePut together algorithms and classes to build your own engineering solutionBuild solvers for industrial optimization problemsDo data analysis using basic and advanced statisticsWho This Book Is For Programmers, data scientists, and analysts with prior experience with programming in any language, especially Java.

This edited volume illustrates the connections between machine learning techniques, black box optimization, and no-free lunch theorems. Each of the thirteen contributions focuses on the commonality and interdisciplinary concepts as well as the fundamentals needed to fully comprehend the impact of individual applications and problems. Current theoretical, algorithmic, and practical methods used are provided to stimulate a new effort towards innovative and efficient solutions. The book is intended for beginners who wish to achieve a broad overview of optimization methods and also for more experienced researchers as well as researchers in mathematics, optimization, operations research, quantitative logistics, data analysis, and statistics, who will benefit from access to a quick reference to key topics and methods. The coverage ranges from mathematically rigorous methods to heuristic and evolutionary approaches in an attempt to equip the reader with different viewpoints of the same problem.

'It's a joy to read a book by a mathematician who knows how to write. [...] There is no better guide to the strategies and stakes of this battle for the future.'---Paul Romer, Nobel Laureate, University Professor in Economics at NYU, and former Chief Economist at the World Bank. “By explaining the flaws and foibles of everything from Google search to QAnon—and by providing level-headed evaluations of efforts to fix them—Noah Giansiracusa offers the perfect starting point for anyone entering the maze of modern digital media.”—Jonathan Rauch, senior fellow at the Brookings Institute and contributing editor of The AtlanticFrom deepfakes to GPT-3, deep learning is now powering a new assault on our ability to tell what's real and what's not, bringing a whole new algorithmic side to fake news. On the other hand, remarkable methods are being developed to help automate fact-checking and the detection of fake news and doctored media. Success in the modern business world requires you to understand these algorithmic currents, and to recognize the strengths, limits, and impacts of deep learning---especially when it comes to discerning the truth and differentiating fact from fiction. This book tells the stories of this algorithmic battle for the truth and how it impacts individuals and society at large. In doing so, it weaves together the human stories and what's at stake here, a simplified technical background on how these algorithms work, and an accessible survey of the research literature exploring these various topics. How Algorithms Create and Prevent Fake News is an accessible, broad account of the various ways that data-driven algorithms have been distorting reality and rendering the truth harder to grasp. From news aggregators to Google searches to YouTube recommendations to Facebook news feeds, the way we obtain information todayis filtered through the lens of tech giant algorithms. The way data is collected, labelled, and stored has a big impact on the machine learning algorithms that are trained on it, and this is a main source of algorithmic bias ­– which gets amplified in harmful data feedback loops. Don't be afraid: with this book you'll see the remedies and technical solutions that are being applied to oppose these harmful trends. There is hope.What You Will LearnThe ways that data labeling and storage impact machine learning and how feedback loops can occurThe history and inner-workings of YouTube's recommendation algorithmThe state-of-the-art capabilities of AI-powered text generation (GPT-3) and video synthesis/doctoring (deepfakes) and how these technologies have been used so farThe algorithmic tools available to help with automated fact-checking and truth-detectionWho This Book is ForPeople who don't have a technical background (in data, computers, etc.) but who would like to learn how algorithms impact society; business leaders who want to know the powers and perils of relying on artificial intelligence. A secondary audience is people with a technical background who want to explore the larger social and societal impact of their work.

Algorithms are now widely employed to make decisions that have increasingly far-reaching impacts on individuals and society as a whole (“algorithmic governance”), which could potentially lead to manipulation, biases, censorship, social discrimination, violations of privacy, property rights, and more. This has sparked a global debate on how to regulate AI and robotics (“governance of algorithms”). This book discusses both of these key aspects: the impact of algorithms, and the possibilities for future regulation.

This book describes in detail sampling techniques that can be used for unsupervised and supervised cases, with a focus on sampling techniques for machine learning algorithms. It covers theory and models of sampling methods for managing scalability and the “curse of dimensionality”, their implementations, evaluations, and applications. A large part of the book is dedicated to database comprising standard feature vectors, and a special section is reserved to the handling of more complex objects and dynamic scenarios. The book is ideal for anyone teaching or learning pattern recognition and interesting teaching or learning pattern recognition and is interested in the big data challenge. It provides an accessible introduction to the field and discusses the state of the art concerning sampling techniques for supervised and unsupervised task.Provides a comprehensive description of sampling techniques for unsupervised and supervised tasks;Describe implementationand evaluation of algorithms that simultaneously manage scalable problems and curse of dimensionality;Addresses the role of sampling in dynamic scenarios, sampling when dealing with complex objects, and new challenges arising from big data.'This book represents a timely collection of state-of-the art research of sampling techniques, suitable for anyone who wants to become more familiar with these helpful techniques for tackling the big data challenge.'M. Emre Celebi, Ph.D., Professor and Chair, Department of Computer Science, University of Central Arkansas'In science the difficulty is not to have ideas, but it is to make them work'From Carlo Rovelli

The recent rapid growth in the variety and complexity of new machine learning architectures requires the development of improved methods for designing, analyzing, evaluating, and communicating machine learning technologies. Statistical Machine Learning: A Unified Framework provides students, engineers, and scientists with tools from mathematical statistics and nonlinear optimization theory to become experts in the field of machine learning. In particular, the material in this text directly supports the mathematical analysis and design of old, new, and not-yet-invented nonlinear high-dimensional machine learning algorithms.Features: Unified empirical risk minimization framework supports rigorous mathematical analyses of widely used supervised, unsupervised, and reinforcement machine learning algorithms Matrix calculus methods for supporting machine learning analysis and design applications Explicit conditions for ensuring convergence of adaptive, batch, minibatch, MCEM, and MCMC learning algorithms that minimize both unimodal and multimodal objective functions Explicit conditions for characterizing asymptotic properties of M-estimators and model selection criteria such as AIC and BIC in the presence of possible model misspecification This advanced text is suitable for graduate students or highly motivated undergraduate students in statistics, computer science, electrical engineering, and applied mathematics. The text is self-contained and only assumes knowledge of lower-division linear algebra and upper-division probability theory. Students, professional engineers, and multidisciplinary scientists possessing these minimal prerequisites will find this text challenging yet accessible.About the Author:Richard M. Golden (Ph.D., M.S.E.E., B.S.E.E.) is Professor of Cognitive Science and Participating Faculty Member in Electrical Engineering at the University of Texas at Dallas. Dr. Golden has published articles and given talks at scientific conferences on a wide range of topics in the fields of both statistics and machine learning over the past three decades. His long-term research interests include identifying conditions for the convergence of deterministic and stochastic machine learning algorithms and investigating estimation and inference in the presence of possibly misspecified probability models.

How can we select the best performing data-driven model? How can we rigorously estimate its generalization error? Statistical learning theory answers these questions by deriving non-asymptotic bounds on the generalization error of a model or, in other words, by upper bounding the true error of the learned model based just on quantities computed on the available data. However, for a long time, Statistical learning theory has been considered only an abstract theoretical framework, useful for inspiring new learning approaches, but with limited applicability to practical problems. The purpose of this book is to give an intelligible overview of the problems of model selection and error estimation, by focusing on the ideas behind the different statistical learning theory approaches and simplifying most of the technical aspects with the purpose of making them more accessible and usable in practice. The book starts by presenting the seminal works of the 80's and includes the most recent results. It discusses open problems and outlines future directions for research.

Introduction to Data Science: Data Analysis and Prediction Algorithms with R introduces concepts and skills that can help you tackle real-world data analysis challenges. It covers concepts from probability, statistical inference, linear regression, and machine learning. It also helps you develop skills such as R programming, data wrangling, data visualization, predictive algorithm building, file organization with UNIX/Linux shell, version control with Git and GitHub, and reproducible document preparation. This book is a textbook for a first course in data science. No previous knowledge of R is necessary, although some experience with programming may be helpful. The book is divided into six parts: R, data visualization, statistics with R, data wrangling, machine learning, and productivity tools. Each part has several chapters meant to be presented as one lecture.The author uses motivating case studies that realistically mimic a data scientist's experience. He starts by asking specific questions and answers these through data analysis so concepts are learned as a means to answering the questions. Examples of the case studies included are: US murder rates by state, self-reported student heights, trends in world health and economics, the impact of vaccines on infectious disease rates, the financial crisis of 2007-2008, election forecasting, building a baseball team, image processing of hand-written digits, and movie recommendation systems. The statistical concepts used to answer the case study questions are only briefly introduced, so complementing with a probability and statistics textbook is highly recommended for in-depth understanding of these concepts. If you read and understand the chapters and complete the exercises, you will be prepared to learn the more advanced concepts and skills needed to become an expert.A complete solutions manual is available to registered instructors who require the text for a course.

Take your Python machine learning ideas and create serverless web applications accessible by anyone with an Internet connection. Some of the most popular serverless cloud providers are covered in this book—Amazon, Microsoft, Google, and PythonAnywhere. You will work through a series of common Python data science problems in an increasing order of complexity. The practical projects presented in this book are simple, clear, and can be used as templates to jump-start many other types of projects. You will learn to create a web application around numerical or categorical predictions, understand the analysis of text, create powerful and interactive presentations, serve restricted access to data, and leverage web plugins to accept credit card payments and donations. You will get your projects into the hands of the world in no time. Each chapter follows three steps: modeling the right way, designing and developing a local web application, and deploying onto a popular and reliable serverless cloud provider. You can easily jump to or skip particular topics in the book. You also will have access to Jupyter notebooks and code repositories for complete versions of the code covered in the book. What You'll Learn Extend your machine learning models using simple techniques to create compelling and interactive web dashboardsLeverage the Flask web framework for rapid prototyping of your Python models and ideasCreate dynamic content powered by regression coefficients, logistic regressions, gradient boosting machines, Bayesian classifications, and moreHarness the power of TensorFlow by exporting saved models into web applicationsCreate rich web dashboards to handle complex real-time user input with JavaScript and Ajax to yield interactive and tailored contentCreate dashboards with paywalls to offer subscription-based accessAccess API data such as Google Maps,OpenWeather, etc.Apply different approaches to make sense of text data and return customized intelligenceBuild an intuitive and useful recommendation site to add value to users and entice them to keep coming backUtilize the freemium offerings of Google Analytics and analyze the resultsTake your ideas all the way to your customer's plate using the top serverless cloud providers Who This Book Is For Those with some programming experience with Python, code editing, and access to an interpreter in working order. The book is geared toward entrepreneurs who want to get their ideas onto the web without breaking the bank, small companies without an IT staff, students wanting exposure and training, and for all data science professionals ready to take things to the next level.

Discover hidden relationships among the variables in your data, and learn how to exploit these relationships. This book presents a collection of data-mining algorithms that are effective in a wide variety of prediction and classification applications. All algorithms include an intuitive explanation of operation, essential equations, references to more rigorous theory, and commented C++ source code.Many of these techniques are recent developments, still not in widespread use. Others are standard algorithms given a fresh look. In every case, the focus is on practical applicability, with all code written in such a way that it can easily be included into any program. The Windows-based DATAMINE program lets you experiment with the techniques before incorporating them into your own work.What You'll LearnUse Monte-Carlo permutation tests to provide statistically sound assessments of relationships present in your dataDiscover how combinatorially symmetric cross validation reveals whether your model has true power or has just learned noise by overfitting the dataWork with feature weighting as regularized energy-based learning to rank variables according to their predictive power when there is too little data for traditional methodsSee how the eigenstructure of a dataset enables clustering of variables into groups that exist only within meaningful subspaces of the dataPlot regions of the variable space where there is disagreement between marginal and actual densities, or where contribution to mutual information is highWho This Book Is ForAnyone interested in discovering and exploiting relationships among variables. Although all code examples are written in C++, the algorithms are described in sufficient detail that they can easily be programmed in any language.

Examine the problem of maintaining the quality of big data and discover novel solutions. You will learn the four V's of big data, including veracity, and study the problem from various angles. The solutions discussed are drawn from diverse areas of engineering and math, including machine learning, statistics, formal methods, and the Blockchain technology. Veracity of Big Data serves as an introduction to machine learning algorithms and diverse techniques such as the Kalman filter, SPRT, CUSUM, fuzzy logic, and Blockchain, showing how they can be used to solve problems in the veracity domain. Using examples, the math behind the techniques is explained in easy-to-understand language.Determining the truth of big data in real-world applications involves using various tools to analyze the available information. This book delves into some of the techniques that can be used. Microblogging websites such as Twitterhave played a major role in public life, including during presidential elections. The book uses examples of microblogs posted on a particular topic to demonstrate how veracity can be examined and established. Some of the techniques are described in the context of detecting veiled attacks on microblogging websites to influence public opinion.What You'll LearnUnderstand the problem concerning data veracity and its ramificationsDevelop the mathematical foundation needed to help minimize the impact of the problem using easy-to-understand language and examplesUse diverse tools and techniques such as machine learning algorithms, Blockchain, and the Kalman filter to address veracity issuesWho This Book Is ForSoftware developers and practitioners, practicing engineers, curious managers, graduate students, and research scholars

This book provides the reader with a basic understanding of the formal concepts of the cluster, clustering, partition, cluster analysis etc. The book explains feature-based, graph-based and spectral clustering methods and discusses their formal similarities and differences. Understanding the related formal concepts is particularly vital in the epoch of Big Data; due to the volume and characteristics of the data, it is no longer feasible to predominantly rely on merely viewing the data when facing a clustering problem. Usually clustering involves choosing similar objects and grouping them together. To facilitate the choice of similarity measures for complex and big data, various measures of object similarity, based on quantitative (like numerical measurement results) and qualitative features (like text), as well as combinations of the two, are described, as well as graph-based similarity measures for (hyper) linked objects and measures for multilayered graphs. Numerous variants demonstrating how such similarity measures can be exploited when defining clustering cost functions are also presented. In addition, the book provides an overview of approaches to handling large collections of objects in a reasonable time. In particular, it addresses grid-based methods, sampling methods, parallelization via Map-Reduce, usage of tree-structures, random projections and various heuristic approaches, especially those used for community detection.

This interdisciplinary reference and guide provides an introduction to modeling methodologies and models which form the starting point for deriving efficient and effective solution techniques, and presents a series of case studies that demonstrate how heuristic and analytical approaches may be used to solve large and complex problems. Topics and features: introduces the key modeling methods and tools, including heuristic and mathematical programming-based models, and queueing theory and simulation techniques; demonstrates the use of heuristic methods to not only solve complex decision-making problems, but also to derive a simpler solution technique; presents case studies on a broad range of applications that make use of techniques from genetic algorithms and fuzzy logic, tabu search, and queueing theory; reviews examples incorporating system dynamics modeling, cellular automata and agent-based simulations, and the use of big data; supplies expanded descriptions and examples in the appendices.

This book summarizes the state-of-the-art in unsupervised learning. The contributors discuss how with the proliferation of massive amounts of unlabeled data, unsupervised learning algorithms, which can automatically discover interesting and useful patterns in such data, have gained popularity among researchers and practitioners. The authors outline how these algorithms have found numerous applications including pattern recognition, market basket analysis, web mining, social network analysis, information retrieval, recommender systems, market research, intrusion detection, and fraud detection. They present how the difficulty of developing theoretically sound approaches that are amenable to objective evaluation have resulted in the proposal of numerous unsupervised learning algorithms over the past half-century. The intended audience includes researchers and practitioners who are increasingly using unsupervised learning algorithms to analyze their data. Topics of interest includeanomaly detection, clustering, feature extraction, and applications of unsupervised learning. Each chapter is contributed by a leading expert in the field.

Presents a detailed study of the major design components that constitute a top-down decision-tree induction algorithm, including aspects such as split criteria, stopping criteria, pruning and the approaches for dealing with missing values. Whereas the strategy still employed nowadays is to use a'generic'decision-tree induction algorithm regardless of the data, the authors argue on the benefits that a bias-fitting strategy could bring to decision-tree induction, in which the ultimate goal is the automatic generation of a decision-tree induction algorithm tailored to the application domain of interest. For such, they discuss how one can effectively discover the most suitable set of components of decision-tree induction algorithms to deal with a wide variety of applications through the paradigm of evolutionary computation, following the emergence of a novel field called hyper-heuristics.'Automatic Design of Decision-Tree Induction Algorithms'would be highly useful for machine learning and evolutionary computation students and researchers alike.

This book provides a comprehensive tutorial on similarity operators. The authors systematically survey the set of similarity operators, primarily focusing on their semantics, while also touching upon mechanisms for processing them effectively.The book starts off by providing introductory material on similarity search systems, highlighting the central role of similarity operators in such systems. This is followed by a systematic categorized overview of the variety of similarity operators that have been proposed in literature over the last two decades, including advanced operators such as RkNN, Reverse k-Ranks, Skyline k-Groups and K-N-Match. Since indexing is a core technology in the practical implementation of similarity operators, various indexing mechanisms are summarized. Finally, current research challenges are outlined, so as to enable interested readers to identify potential directions for future investigations. In summary, this book offers a comprehensive overview of the field of similarity search operators, allowing readers to understand the area of similarity operators as it stands today, and in addition providing them with the background needed to understand recent novel approaches.

Data Mining Algorithms is a practical, technically-oriented guide to data mining algorithms that covers the most important algorithms for building classification, regression, and clustering models, as well as techniques used for attribute selection and transformation, model quality evaluation, and creating model ensembles. The author presents many of the important topics and methodologies widely used in data mining, whilst demonstrating the internal operation and usage of data mining algorithms using examples in R.

In the fields of data mining and control, the huge amount of unstructured data and the presence of uncertainty in system descriptions have always been critical issues. The book Randomized Algorithms in Automatic Control and Data Mining introduces the readers to the fundamentals of randomized algorithm applications in data mining (especially clustering) and in automatic control synthesis. The methods proposed in this book guarantee that the computational complexity of classical algorithms and the conservativeness of standard robust control techniques will be reduced. It is shown that when a problem requires'brute force'in selecting among options, algorithms based on random selection of alternatives offer good results with certain probability for a restricted time and significantly reduce the volume of operations.

This Springer Brief presents a basic algorithm that provides a correct solution to finding an optimal state change attempt, as well as an enhanced algorithm that is built on top of the well-known trie data structure. It explores correctness and algorithmic complexity results for both algorithms and experiments comparing their performance on both real-world and synthetic data. Topics addressed include optimal state change attempts, state change effectiveness, different kind of effect estimators, planning under uncertainty and experimental evaluation. These topics will help researchers analyze tabular data, even if the data contains states (of the world) and events (taken by an agent) whose effects are not well understood. Event DBs are omnipresent in the social sciences and may include diverse scenarios from political events and the state of a country to education-related actions and their effects on a school system. With a wide range of applications in computer science and the social sciences, the information in this Springer Brief is valuable for professionals and researchers dealing with tabular data, artificial intelligence and data mining. The applications are also useful for advanced-level students of computer science.

This SpringerBrief addresses the challenges of analyzing multi-relational and noisy data by proposing several Statistical Relational Learning (SRL) methods. These methods combine the expressiveness of first-order logic and the ability of probability theory to handle uncertainty. It provides an overview of the methods and the key assumptions that allow for adaptation to different models and real world applications. The models are highly attractive due to their compactness and comprehensibility but learning their structure is computationally intensive. To combat this problem, the authors review the use of functional gradients for boosting the structure and the parameters of statistical relational models. The algorithms have been applied successfully in several SRL settings and have been adapted to several real problems from Information extraction in text to medical problems. Including both context and well-tested applications, Boosting Statistical Relational Learning from Benchmarks to Data-Driven Medicine is designed for researchers and professionals in machine learning and data mining. Computer engineers or students interested in statistics, data management, or health informatics will also find this brief a valuable resource.

Some probability problems are so difficult that they stump the smartest mathematicians. But even the hardest of these problems can often be solved with a computer and a Monte Carlo simulation, in which a random-number generator simulates a physical process, such as a million rolls of a pair of dice. This is what Digital Dice is all about: how to get numerical answers to difficult probability problems without having to solve complicated mathematical equations. Popular-math writer Paul Nahin challenges readers to solve twenty-one difficult but fun problems, from determining the odds of coin-flipping games to figuring out the behavior of elevators. Problems build from relatively easy (deciding whether a dishwasher who breaks most of the dishes at a restaurant during a given week is clumsy or just the victim of randomness) to the very difficult (tackling branching processes of the kind that had to be solved by Manhattan Project mathematician Stanislaw Ulam). In his characteristic style, Nahin brings the problems to life with interesting and odd historical anecdotes. Readers learn, for example, not just how to determine the optimal stopping point in any selection process but that astronomer Johannes Kepler selected his second wife by interviewing eleven women. The book shows readers how to write elementary computer codes using any common programming language, and provides solutions and line-by-line walk-throughs of a MATLAB code for each problem. Digital Dice will appeal to anyone who enjoys popular math or computer science. In a new preface, Nahin wittily addresses some of the responses he received to the first edition.

Random Generation of Trees is about a field on the crossroads between computer science, combinatorics and probability theory. Computer scientists need random generators for performance analysis, simulation, image synthesis, etc. In this context random generation of trees is of particular interest. The algorithms presented here are efficient and easy to code. Some aspects of Horton--Strahler numbers, programs written in C and pictures are presented in the appendices. The complexity analysis is done rigorously both in the worst and average cases. Random Generation of Trees is intended for students in computer science and applied mathematics as well as researchers interested in random generation.

This work presents a data visualization technique that combines graph-based topology representation and dimensionality reduction methods to visualize the intrinsic data structure in a low-dimensional vector space. The application of graphs in clustering and visualization has several advantages. A graph of important edges (where edges characterize relations and weights represent similarities or distances) provides a compact representation of the entire complex data set. This text describes clustering and visualization methods that are able to utilize information hidden in these graphs, based on the synergistic combination of clustering, graph-theory, neural networks, data visualization, dimensionality reduction, fuzzy methods, and topology learning. The work contains numerous examples to aid in the understanding and implementation of the proposed algorithms, supported by a MATLAB toolbox available at an associated website.

The proceedings of the 5th International Workshop on Parallel Tools for High Performance Computing provide an overview on supportive software tools and environments in the fields of System Management, Parallel Debugging and Performance Analysis. In the pursuit to maintain exponential growth for the performance of high performance computers the HPC community is currently targeting Exascale Systems. The initial planning for Exascale already started when the first Petaflop system was delivered. Many challenges need to be addressed to reach the necessary performance. Scalability, energy efficiency and fault-tolerance need to be increased by orders of magnitude. The goal can only be achieved when advanced hardware is combined with a suitable software stack. In fact, the importance of software is rapidly growing. As a result, many international projects focus on the necessary software.

​​​​​​ Stationarity and Convergence in Reduce-or-Retreat Minimization presents and analyzes a unifying framework for a wide variety of numerical methods in optimization. The author's “reduce-or-retreat” framework is a conceptual method-outline that covers any method whose iterations choose between reducing the objective in some way at a trial point, or retreating to a closer set of trial points. The alignment of various derivative-based methods within the same framework encourages the construction of new methods, and inspires new theoretical developments as companions to results from across traditional divides. The text illustrates the former by developing two generalizations of classic derivative-based methods which accommodate non-smooth objectives, and the latter by analyzing these two methods in detail along with a pattern-search method and the famous Nelder-Mead method.In addition to providing a bridge for theory through the “reduce-or-retreat” framework, this monograph extends and broadens the traditional convergence analyses in several ways. Levy develops a generalized notion of approaching stationarity which applies to non-smooth objectives, and explores the roles of the descent and non-degeneracy conditions in establishing this property. The traditional analysis is broadened by considering “situational” convergence of different elements computed at each iteration of a reduce-or-retreat method. The “reduce-or-retreat” framework described in this text covers specialized minimization methods, some general methods for minimization and a direct search method, while providing convergence analysis which complements and expands existing results.​ ​

Due to the fast growth of the Web and the difficulties in finding desired information, efficient and effective information retrieval systems have become more important than ever, and the search engine has become an essential tool for many people. The ranker, a central component in every search engine, is responsible for the matching between processed queries and indexed documents. Because of its central role, great attention has been paid to the research and development of ranking technologies. In addition, ranking is also pivotal for many other information retrieval applications, such as collaborative filtering, definition ranking, question answering, multimedia retrieval, text summarization, and online advertisement. Leveraging machine learning technologies in the ranking process has led to innovative and more effective ranking models, and eventually to a completely new research area called “learning to rank”. Liu first gives a comprehensive review of the major approaches to learning to rank. For each approach he presents the basic framework, with example algorithms, and he discusses its advantages and disadvantages. He continues with some recent advances in learning to rank that cannot be simply categorized into the three major approaches – these include relational ranking, query-dependent ranking, transfer ranking, and semisupervised ranking. His presentation is completed by several examples that apply these technologies to solve real information retrieval problems, and by theoretical discussions on guarantees for ranking performance. This book is written for researchers and graduate students in both information retrieval and machine learning. They will find here the only comprehensive description of the state of the art in a field that has driven the recent advances in search engine development.

While many think of algorithms as specific to computer science, at its core algorithmic thinking is defined by the use of analytical logic to solve problems. This logic extends far beyond the realm of computer science and into the wide and entertaining world of puzzles. In Algorithmic Puzzles, Anany and Maria Levitin use many classic brainteasers as well as newer examples from job interviews with major corporations to show readers how to apply analytical thinking to solve puzzles requiring well-defined procedures. The book's unique collection of puzzles is supplemented with carefully developed tutorials on algorithm design strategies and analysis techniques intended to walk the reader step-by-step through the various approaches to algorithmic problem solving. Mastery of these strategies--exhaustive search, backtracking, and divide-and-conquer, among others--will aid the reader in solving not only the puzzles contained in this book, but also others encountered in interviews, puzzle collections, and throughout everyday life. Each of the 150 puzzles contains hints and solutions, along with commentary on the puzzle's origins and solution methods. The only book of its kind, Algorithmic Puzzles houses puzzles for all skill levels. Readers with only middle school mathematics will develop their algorithmic problem-solving skills through puzzles at the elementary level, while seasoned puzzle solvers will enjoy the challenge of thinking through more difficult puzzles.

In operations research and computer science it is common practice to evaluate the performance of optimization algorithms on the basis of computational results, and the experimental approach should follow accepted principles that guarantee the reliability and reproducibility of results. However, computational experiments differ from those in other sciences, and the last decade has seen considerable methodological research devoted to understanding the particular features of such experiments and assessing the related statistical methods. This book consists of methodological contributions on different scenarios of experimental analysis. The first part overviews the main issues in the experimental analysis of algorithms, and discusses the experimental cycle of algorithm development; the second part treats the characterization by means of statistical distributions of algorithm performance in terms of solution quality, runtime and other measures; and the third part collects advanced methods from experimental design for configuring and tuning algorithms on a specific class of instances with the goal of using the least amount of experimentation. The contributor list includes leading scientists in algorithm design, statistical design, optimization and heuristics, and most chapters provide theoretical background and are enriched with case studies. This book is written for researchers and practitioners in operations research and computer science who wish to improve the experimental assessment of optimization algorithms and, consequently, their design.

This book presents the first cohesive treatment of Information Theoretic Learning (ITL) algorithms to adapt linear or nonlinear learning machines both in supervised or unsupervised paradigms. ITL is a framework where the conventional concepts of second order statistics (covariance, L2 distances, correlation functions) are substituted by scalars and functions with information theoretic underpinnings, respectively entropy, mutual information and correntropy. ITL quantifies the stochastic structure of the data beyond second order statistics for improved performance without using full-blown Bayesian approaches that require a much larger computational cost. This is possible because of a non-parametric estimator of Renyi's quadratic entropy that is only a function of pairwise differences between samples. The book compares the performance of ITL algorithms with the second order counterparts in many engineering and machine learning applications.Students, practitioners and researchers interested in statistical signal processing, computational intelligence, and machine learning will find in this book the theory to understand the basics, the algorithms to implement applications, and exciting but still unexplored leads that will provide fertile ground for future research. José C. Principe is Distinguished Professor of Electrical and Biomedical Engineering, and BellSouth Professor at the University of Florida, and the Founder and Director of the Computational NeuroEngineering Laboratory. He is an IEEE and AIMBE Fellow, Past President of the International Neural Network Society, Past Editor-in-Chief of the IEEE Trans. on Biomedical Engineering and the Founder Editor-in-Chief of the IEEE Reviews on Biomedical Engineering. He has written an interactive electronic book on Neural Networks, a book on Brain Machine Interface Engineering and more recently a book on Kernel Adaptive Filtering, and was awarded the 2011 IEEE NeuralNetwork Pioneer Award.

Data mining is a very active research area with many successful real-world app- cations. It consists of a set of concepts and methods used to extract interesting or useful knowledge (or patterns) from real-world datasets, providing valuable support for decision making in industry, business, government, and science. Although there are already many types of data mining algorithms available in the literature, it is still dif cult for users to choose the best possible data mining algorithm for their particular data mining problem. In addition, data mining al- rithms have been manually designed; therefore they incorporate human biases and preferences. This book proposes a new approach to the design of data mining algorithms. - stead of relying on the slow and ad hoc process of manual algorithm design, this book proposes systematically automating the design of data mining algorithms with an evolutionary computation approach. More precisely, we propose a genetic p- gramming system (a type of evolutionary computation method that evolves c- puter programs) to automate the design of rule induction algorithms, a type of cl- si cation method that discovers a set of classi cation rules from data. We focus on genetic programming in this book because it is the paradigmatic type of machine learning method for automating the generation of programs and because it has the advantage of performing a global search in the space of candidate solutions (data mining algorithms in our case), but in principle other types of search methods for this task could be investigated in the future.

Interval computing combined with fuzzy logic has become an emerging tool in studying artificial intelligence and knowledge processing (AIKP) applications since it models uncertainties frequently raised in the field. This book provides introductions for both interval and fuzzy computing in a very accessible style. Application algorithms covered in this book include quantitative and qualitative data mining with interval valued datasets, decision making systems with interval valued parameters, interval valued Nash games and interval weighted graphs. Successful applications in studying finance and economics, etc are also included. This book can serve as a handbook or a text for readers interested in applying interval and soft computing for AIKP.

Data Streams: Models and Algorithms primarily discusses issues related to the mining aspects of data streams. Recent progress in hardware technology makes it possible for organizations to store and record large streams of transactional data. For example, even simple daily transactions such as using the credit card or phone result in automated data storage, which brings us to a fairly new topic called data streams. This volume covers mining aspects of data streams comprehensively: each contributed chapter contains a survey on the topic, the key ideas in the field for that particular topic, and future research directions. Data Streams: Models and Algorithms is intended for a professional audience composed of researchers and practitioners in industry. This book is also appropriate for advanced-level students in computer science.

There is a growing need for a more automated system of partitioning data sets into groups, or clusters. For example, digital libraries and the World Wide Web continue to grow exponentially, the ability to find useful information increasingly depends on the indexing infrastructure or search engine. Clustering techniques can be used to discover natural groups in data sets and to identify abstract structures that might reside there, without having any background knowledge of the characteristics of the data. Clustering has been used in a variety of areas, including computer vision, VLSI design, data mining, bio-informatics (gene expression analysis), and information retrieval, to name just a few. This book focuses on a few of the most important clustering algorithms, providing a detailed account of these major models in an information retrieval context. The beginning chapters introduce the classic algorithms in detail, while the later chapters describe clustering through divergences and show recent research for more advanced audiences.

Over the last few decades, important progresses in the methods of sampling have been achieved. This book draws up an inventory of new methods that can be useful for selecting samples. Forty-six sampling methods are described in the framework of general theory. The algorithms are described rigorously, which allows implementing directly the described methods. This book is aimed at experienced statisticians who are familiar with the theory of survey sampling.

In many applications, e.g., bioinformatics, web access traces, system u- lization logs, etc., the data is naturally in the form of sequences. It has been of great interests to analyze the sequential data to find their inherent char- teristics. The sequential pattern is one of the most widely studied models to capture such characteristics. Examples of sequential patterns include but are not limited to protein sequence motifs and web page navigation traces. In this book, we focus on sequential pattern mining. To meet different needs of various applications, several models of sequential patterns have been proposed. We do not only study the mathematical definitions and application domains of these models, but also the algorithms on how to effectively and efficiently find these patterns. The objective of this book is to provide computer scientists and domain - perts such as life scientists with a set of tools in analyzing and understanding the nature of various sequences by : (1) identifying the specific model(s) of - quential patterns that are most suitable, and (2) providing an efficient algorithm for mining these patterns. Chapter 1 INTRODUCTION Data Mining is the process of extracting implicit knowledge and discovery of interesting characteristics and patterns that are not explicitly represented in the databases. The techniques can play an important role in understanding data and in capturing intrinsic relationships among data instances. Data mining has been an active research area in the past decade and has been proved to be very useful.

Written for practitioners of data mining, data cleaning and database management. Presents a technical treatment of data quality including process, metrics, tools and algorithms. Focuses on developing an evolving modeling strategy through an iterative data exploration loop and incorporation of domain knowledge. Addresses methods of detecting, quantifying and correcting data quality issues that can have a significant impact on findings and decisions, using commercially available tools as well as new algorithmic approaches. Uses case studies to illustrate applications in real life scenarios. Highlights new approaches and methodologies, such as the DataSphere space partitioning and summary based analysis techniques. Exploratory Data Mining and Data Cleaning will serve as an important reference for serious data analysts who need to analyze large amounts of unfamiliar data, managers of operations databases, and students in undergraduate or graduate level courses dealing with large scale data analys is and data mining.

The term “stringology” is a popular nickname for text algorithms, or algorithms on strings. This book deals with the most basic algorithms in the area. Most of them can be viewed as “algorithmic jewels” and deserve reader-friendly presentation. One of the main aims of the book is to present several of the most celebrated algorithms in a simple way by omitting obscuring details and separating algorithmic structure from combinatorial theoretical background. The book reflects the relationships between applications of text-algorithmic techniques and the classification of algorithms according to the measures of complexity considered. The text can be viewed as a parade of algorithms in which the main purpose is to discuss the foundations of the algorithms and their interconnections. One can partition the algorithmic problems discussed into practical and theoretical problems. Certainly, string matching and data compression are in the former class, while most problems related to symmetries and repetitions in texts are in the latter. However, all the problems are interesting from an algorithmic point of view and enable the reader to appreciate the importance of combinatorics on words as a tool in the design of efficient text algorithms.In most textbooks on algorithms and data structures, the presentation of efficient algorithms on words is quite short as compared to issues in graph theory, sorting, searching, and some other areas. At the same time, there are many presentations of interesting algorithms on words accessible only in journals and in a form directed mainly at specialists. This book fills the gap in the book literature on algorithms on words, and brings together the many results presently dispersed in the masses of journal articles. The presentation is reader-friendly; many examples and about two hundred figures illustrate nicely the behaviour of otherwise very complex algorithms.