||New Reviews| |Software Methodologies| |Popular Science| |AI/Machine Learning| |Programming| |Java| |Linux/Open Source| |XML| |Software Tools| |Other| |Web| |Tutorials| |All By Date| |All By Title| |Resources| |About||
Keywords: Machine learning, statistics, genetic algorithms, neural networks, data analysis
Title:Intelligent Data Analysis: An IntroductionEditors: Michael Berthold and David J. Hand
Verdict: Academically rigorous and comprehensive survey of the field of Intelligent Data Analysis
Intelligent Data Analysis (IDA) is an emerging discipline that is at the intersection of statistical analysis, machine learning, data mining and knowledge discovery. Driven by the increasingly sophisticated requirements of analysis vast amounts of data, it throws up many challenges that traditional data analysis methods are unable to deal with in isolation. In this second edition of 'Intelligent Data Analysis: An Introduction', editors Michael Berthold and David Hand have expanded the coverage of topics and ensured that this remains the key text for surveying the field.
The twelve chapters which make up the book provide an academically rigorous and concise to the key methodologies which make up the discipline. Beginning with classical statistical concepts, from introductory material on probability onwards through to inference, prediction, multi-variate statistics, Bayesian methods and models.
Other chapters look at Support Vector and Kernel Methods, Analysis of Time Series (including non-linear dynamics), Rule Induction (inductive logic programming), Neural Networks, Fuzzy Logic, Stochastic Search (simulated annealing and evolutionary algorithms), Visualisation and the final chapter covers Systems and Applications.
Each of the chapters can serve as a fully referenced survey of a given area, touching not just introductory material but also more advanced topics which pertain to data analysis. The neural networks chapter, for example, begins with the basic perceptron and moves quickly into multi-layer feedforward networks, radial basis function networks and competitive learning. Applications of neural networks are not ignored, with principal component and time series analysis examined in more detail.
Similarly the chapter on Stochastic Search Methods examines a full range of methodologies, including simulated annealing, evolution strategies, genetic algorithms and genetic programming. The treatment is rigorous and mathematical, indeed the book emphasises solid theoretical foundations above empirical or implementation issues.
The final chapter looks at practical applications of intelligent analysis, giving a broad over-view of the types of tools being developed and deployed across a range of problem domains.
In all this is a comprehensive survey of the field, and will appeal to graduate and post-graduate students, researchers and academics seeking an overview of the theoretical tools available for intelligently analysing large, complex data sets.