Selected Contributions in Data Analysis and Classification

1. Front Matter

   Pages I-XIII

   PDF

2. Analysis of Symbolic Data

   1. Front Matter

      Pages 1-1

      PDF

   2. Dependencies and Variation Components of Symbolic Interval-Valued Data

      • Lynne Billard

      Pages 3-12

   3. On the Analysis of Symbolic Data

      • Paula Brito

      Pages 13-22

   4. Symbolic Analysis to Learn Evolving CyberTraffic

      • Costantina Caruso, Donato Malerba

      Pages 23-33

   5. A Clustering Algorithm for Symbolic Interval Data Based on a Single Adaptive Hausdorff Distance

      • Francisco de A. T. de Carvalho

      Pages 35-44

   6. An Agglomerative Hierarchical Clustering Algorithm for Improving Symbolic Object Retrieval

      • Floriana Esposito, Claudia d’Amato

      Pages 45-53

   7. 3WaySym-Scal: Three-Way Symbolic Multidimensional Scaling

      • Patrick J. F. Groenen, Suzanne Winsberg

      Pages 55-67

   8. Clustering and Validation of Interval Data

      • André Hardy, Joffray Baune

      Pages 69-81

   9. Building Symbolic Objects from Data Streams

      • Georges Hébrail, Yves Lechevallier

      Pages 83-94

   10. Feature Clustering Method to Detect Monotonic Chain Structures in Symbolic Data

       • Manabu Ichino

       Pages 95-102

   11. Symbolic Markov Chains

       • Monique Noirhomme-Fraiture, Etienne Cuvelier

       Pages 103-111

   12. Quality Issues in Symbolic Data Analysis

       • Haralambos Papageorgiou, Maria Vardaki

       Pages 113-122

   13. Dynamic Clustering of Histogram Data: Using the Right Metric

       • Rosanna Verde, Antonio Irpino

       Pages 123-134

3. Clustering Methods

   1. Front Matter

      Pages 135-135

      PDF

   2. Beyond the Pyramids: Rigid Clustering Systems

      • Jean-Pierre Barthélemy, Gentian Gusho, Christophe Osswald

      Pages 137-150

   3. Indirect Blockmodeling of 3-Way Networks

      • Vladimir Batagelj, Anuška Ferligoj, Patrick Doreian

      Pages 151-159

   4. Clustering Methods: A History of k-Means Algorithms

      • Hans-Hermann Bock

      Pages 161-172

   5. Overlapping Clustering in a Graph Using k-Means and Application to Protein Interactions Networks

      • Irène Charon, Lucile Denoeud, Olivier Hudry

      Pages 173-182

   6. Species Clustering via Classical and Interval Data Representation

      • Marie Chavent

      Pages 183-191

By inviting me to write a preface, the organizers of the event in honour of Edwin Diday, have expressed their a?ection and I appreciate this very much. This gives me an opportunity to express my friendship and admiration for Edwin Diday, and I wrote this foreword with pleasure. My ?rst few meetings withEdwinDidaydatebackto1965through1975,daysofthedevelopmentof French statistics. This was a period when access to computers revolutionized the practice of statistics. This does not refer to individual computers or to terminals that have access to powerful networks. This was the era of the ?rst university calculation centres that one accessed over a counter. One would deposit cards on which program and data were punched in and come back a few hours or days later for the results. Like all those who used linear data analysis, the computer enabled me to calculate for each data set the value of mathematical objects (eigenvalues and eigenvectors for example) whose optimality properties had been demonstrated by mathematicians. It was - ready a big step to be able to do this in concrete experimental situations. With Dynamic Clustering Algorithm, Edwin Diday allowed us to discover that computers could be more than just a way of giving numerical values to known mathematical objects. Besides the e?ciency of the solutions he built, he led us to integrate the access to computers di?erently in the research and practice of data analysis.

• Random variable
• algorithms
• classification
• cluster analysis
• clustering
• data analysis
• data mining
• knowledge discovery
• knowledge management
• learning
• multivariate statistics
• operations research
• optimization
• principal component analysis
• statistics

• Faculty of Economics, University of Porto, Porto, Portugal

  Paula Brito

• ESG UQAM, Montreal, Canada

  Guy Cucumel

• Department Lussi, ENST Bretagne, Cesson-Sévigné Cedex, France

  Patrice Bertrand

• Centre of Computer Science (CIn), Federal University of Pernambuco (UFPE), Recife-PE, Brazil

  Francisco Carvalho

Policies and ethics