Data Science and Classification

ISBN-13: 9783540344155 / Angielski / Miękka / 2006 / 358 str.

Vladimir Batagelj; Hans-Hermann Bock; Anue!ka Ferligoj

Opis

Data Science and Classification provides new methodological developments in data analysis and classification. The broad and comprehensive coverage includes the measurement of similarity and dissimilarity, methods for classification and clustering, network and graph analyses, analysis of symbolic data, and web mining. Beyond structural and theoretical results, the book offers application advice for a variety of problems, in medicine, microarray analysis, social network structures, and music.

Recenzje

From the reviews:

"This book is a collection of papers presented at the Tenth Conference of the International Federation of Classification Societies. The contributors are primarily statisticians and computer scientists ... . The typesetting and page layout are well done, and the graphics are very clear. ... The main market for this book would be libraries, and researchers wanting a record of recent advances in statistical learning." (Jeffrey D. Picka, Technometrics, Vol. 49 (3), August, 2007)

Spis treści

Similarity and Dissimilarity.- A Tree-Based Similarity for Evaluating Concept Proximities in an Ontology.- Improved Fréchet Distance for Time Series.- Comparison of Distance Indices Between Partitions.- Design of Dissimilarity Measures: A New Dissimilarity Between Species Distribution Areas.- Dissimilarities for Web Usage Mining.- Properties and Performance of Shape Similarity Measures.- Classification and Clustering.- Hierarchical Clustering for Boxplot Variables.- Evaluation of Allocation Rules Under Some Cost Constraints.- Crisp Partitions Induced by a Fuzzy Set.- Empirical Comparison of a Monothetic Divisive Clustering Method with the Ward and the k-means Clustering Methods.- Model Selection for the Binary Latent Class Model: A Monte Carlo Simulation.- Finding Meaningful and Stable Clusters Using Local Cluster Analysis.- Comparing Optimal Individual and Collective Assessment Procedures.- Network and Graph Analysis.- Some Open Problem Sets for Generalized Blockmodeling.- Spectral Clustering and Multidimensional Scaling: A Unified View.- Analyzing the Structure of U.S. Patents Network.- Identifying and Classifying Social Groups: A Machine Learning Approach.- Analysis of Symbolic Data.- Multidimensional Scaling of Histogram Dissimilarities.- Dependence and Interdependence Analysis for Interval-Valued Variables.- A New Wasserstein Based Distance for the Hierarchical Clustering of Histogram Symbolic Data.- Symbolic Clustering of Large Datasets.- A Dynamic Clustering Method for Mixed Feature-Type Symbolic Data.- General Data Analysis Methods.- Iterated Boosting for Outlier Detection.- Sub-species of Homopus Areolatus? Biplots and Small Class Inference with Analysis of Distance.- Revised Boxplot Based Discretization as the Kernel of Automatic Interpretation of Classes Using Numerical Variables.- Data and Web Mining.- Comparison of Two Methods for Detecting and Correcting Systematic Error in High-throughput Screening Data.- kNN Versus SVM in the Collaborative Filtering Framework.- Mining Association Rules in Folksonomies.- Empirical Analysis of Attribute-Aware Recommendation Algorithms with Variable Synthetic Data.- Patterns of Associations in Finite Sets of Items.- Analysis of Music Data.- Generalized N-gram Measures for Melodic Similarity.- Evaluating Different Approaches to Measuring the Similarity of Melodies.- Using MCMC as a Stochastic Optimization Procedure for Musical Time Series.- Local Models in Register Classification by Timbre.- Gene and Microarray Analysis.- Improving the Performance of Principal Components for Classification of Gene Expression Data Through Feature Selection.- A New Efficient Method for Assessing Missing Nucleotides in DNA Sequences in the Framework of a Generic Evolutionary Model.- New Efficient Algorithm for Modeling Partial and Complete Gene Transfer Scenarios.