Bock H.-H., Diday E. (eds.) Analysis of Symbolic Data. Exploratory Methods for Extracting Statistical Information from Complex Data

Springer, 2000. — 444.This book presents statistical methods for analyzing and visualizing complex, aggregated, relational and higher-level data. It extends classical methods of exploratory data analysis to the case of so-called 'symbolic variables' that may have values such as subsets of categories, intervals of the real line (e.g., minimum/ maximum daily sales), or frequency distributions (such as the income distribution in different countries), which is in contrast to the classical approach where only one single number, category or level is allowed. In addition, logical relationships may be defined between different variables, either by rules or by taxonomies. Since classical data are a special case of symbolic data, it results that classical exploratory data analysis is a special case of the Symbolic Data Analysis (SDA) presented in this book.Symbolic Data Analysis and the SODAS Project: Purpose, History, Perspective
The Classical Data Situation
Symbolic Data
Symbolic Objects
Generation of Symbolic Objects from Relational Databases
Descriptive Statistics for Symbolic Data
Visualizing and Editing Symbolic Objects
Similarity and Dissimilarity
Symbolic Factor Analysis
Discrimination: Assigning Symbolic Objects to Classes
Clustering Methods for Symbolic Objects
Symbolic Approaches for Three-way Data
Illustrative Benchmark Analyses
The SODAS Software Package