New Developments in Classification and Data Analysis

1. Front Matter

   Pages I-X

   PDF

2. Classification and Clustering

   1. Multi-Class Budget Exploratory Trees

      • Massimo Aria

      Pages 3-10

   2. Methods to Compare Nonparametric Classifiers and to Select the Predictors

      • Simone Borra, Agostino Di Ciaccio

      Pages 11-19

   3. Variable Selection in Cell Classification Problems: A Strategy Based on Independent Component Analysis

      • Daniela G. Calò, Giuliano Galimberti, Marilena Pillati, Cinzia Viroli

      Pages 21-29

   4. Simplifying Classification Trees Through Consensus Methods

      • Rossella Miglio, Gabriele Soffritti

      Pages 31-37

   5. Selecting the Training Set in Classification Problems with Rare Events

      • Bruno Scarpa, Nicola Torelli

      Pages 39-46

   6. A Classification and Discrimination Integrated Strategy Conducted on Symbolic Data for Missing Data Treatment in Questionnaire Survey

      • Maria Gabriella Grassia

      Pages 47-54

   7. A Collinearity Based Hierarchical Method to Identify Clusters of Variables

      • Annalisa Laghi, Gabriele Soffritti

      Pages 55-62

   8. On the Dynamic Time Warping for Computing the Dissimilarity Between Curves

      • Isabella Morlini

      Pages 63-70

   9. Metrics in Symbolic Data Analysis

      • Luciano Nieddu, Alfredo Rizzi

      Pages 71-78

   10. Constrained Clusterwise Linear Regression

       • Antonella Plaia

       Pages 79-86

   11. Crossed Clustering Method on Symbolic Data Tables

       • Rosanna Verde, Yves Lechevallier

       Pages 87-94

3. Multivariate Statistics and Data Analysis

   1. Some Statistical Applications of Centrosymmetric Matrices

      • Estela Bee Dagum, Laura Guidotti, Alessandra Luati

      Pages 97-104

   2. Selection of Structural Equation Models with the PLS-VB Programme

      • Giuseppe Boari, Gabriele Cantaluppi

      Pages 105-112

   3. Web Robot Detection - Preprocessing Web Logfiles for Robot Detection

      • Christian Bomhardt, Wolfgang Gaul, Lars Schmidt-Thieme

      Pages 113-124

   4. Data Dependent Prior Modeling and Estimation in Contingency Tables: The Order-Restricted RC Model

      • Giulio D'Epifanio

      Pages 125-132

   5. PLS Typological Regression: Algorithmic, Classification and Validation Issues

      • Vincenzo Esposito Vinzi, Carlo N. Lauro, Silvano Amato

      Pages 133-140

   6. Weighted Metric Multidimensional Scaling

      • Michael Greenacre

      Pages 141-149

   7. An Improved Majorization Algorithm for Robust Procrustes Analysis

      • Patrick J.F. Groenen, Patrizia Giaquinto, Henk A.L. Kiers

      Pages 151-158

   8. Generalized Bi-Additive Modelling for Categorical Data

      • Patrick J.F. Groenen, Alex J. Koning

      Pages 159-166

This volume contains revised versions of selected papers presented during the biannual meeting of the Classification and Data Analysis Group of SocietA Italiana di Statistica, which was held in Bologna, September 22-24, 2003. The scientific program of the conference included 80 contributed papers. Moreover it was possible to recruit six internationally renowned invited spe- ers for plenary talks on their current research works regarding the core topics of IFCS (the International Federation of Classification Societies) and Wo- gang Gaul and the colleagues of the GfKl organized a session. Thus, the conference provided a large number of scientists and experts from home and abroad with an attractive forum for discussions and mutual exchange of knowledge. The talks in the different sessions focused on methodological developments in supervised and unsupervised classification and in data analysis, also p- viding relevant contributions in the context of applications. This suggested the presentation of the 43 selected papers in three parts as follows: CLASSIFICATION AND CLUSTERING Non parametric classification Clustering and dissimilarities MULTIVARIATE STATISTICS AND DATA ANALYSIS APPLIED MULTIVARIATE STATISTICS Environmental data Microarray data Behavioural and text data Financial data We wish to express our gratitude to the authors whose enthusiastic p- ticipation made the meeting possible. We are very grateful to the reviewers for the time spent in their professional reviewing work. We would also like to extend our thanks to the chairpersons and discussants of the sessions: their comments and suggestions proved very stimulating both for the authors and the audience.

• Access
• Multivariate statistics
• Time series
• calculus
• classification
• cluster analysis
• clustering
• data analysis
• data mining
• linear regression
• modeling
• multidimensional scaling
• performance
• principal component analysis
• statistics

• Aachen

  H.-H. Bock

• Karlsruhe

  W. Gaul

• Rome

  M. Vichi

• Newark

  Ph. Arabie

• Cottbus

  D. Baier

• Milton Keynes

  F. Critchley

• Bielefeld

  R. Decker

• Paris

  E. Diday

• Barcelona

  M. Greenacre

• Naples

  C. Lauro

• Leiden

  J. Meulman

• Bologna

  P. Monari

• Toronto

  S. Nishisato

• Tokyo

  N. Ohsumi

• Augsburg

  O. Opitz

• Passau

  G. Ritter

• Mannheim

  M. Schader

• Dortmund

  C. Weihs

• Department of Statistics, Probability and Applied Statistics, University of Rome “La Sapienza”, Rome, Italy

  Maurizio Vichi

• Department of Statistical Sciences, University of Bologna, Bologna, Italy

  Paola Monari, Stefania Mignani, Angela Montanari

Policies and ethics