Preface |
List of Denotations |
Introduction: Historical Remarks |
What Is Clustering / 1: |
Base words |
Exemplary problems / 1.1: |
Structuring / 1.1.1: |
Description / 1.1.2: |
Association / 1.1.3: |
Generalization / 1.1.4: |
Visualization of data structure / 1.1.5: |
Bird's-eye view / 1.2: |
Definition: data and cluster structure / 1.2.1: |
Criteria for revealing a cluster structure / 1.2.2: |
Three types of cluster description / 1.2.3: |
Stages of a clustering application / 1.2.4: |
Clustering and other disciplines / 1.2.5: |
Different perspectives of clustering / 1.2.6: |
What Is Data / 2: |
Feature characteristics / 2.1: |
Feature scale types / 2.1.1: |
Quantitative case / 2.1.2: |
Categorical case / 2.1.3: |
Bivariate analysis / 2.2: |
Two quantitative variables / 2.2.1: |
Nominal and quantitative variables / 2.2.2: |
Two nominal variables cross-classified / 2.2.3: |
Relation between correlation and contingency / 2.2.4: |
Meaning of correlation / 2.2.5: |
Feature space and data scatter / 2.3: |
Data matrix / 2.3.1: |
Feature space: distance and inner product / 2.3.2: |
Data scatter / 2.3.3: |
Pre-processing and standardizing mixed data / 2.4: |
Other table data types / 2.5: |
Dissimilarity and similarity data / 2.5.1: |
Contingency and flow data / 2.5.2: |
K-Means Clustering / 3: |
Conventional K-Means / 3.1: |
Straight K-Means / 3.1.1: |
Square error criterion / 3.1.2: |
Incremental versions of K-Means / 3.1.3: |
Initialization of K-Means / 3.2: |
Traditional approaches to initial setting / 3.2.1: |
MaxMin for producing deviate centroids / 3.2.2: |
Deviate centroids with Anomalous pattern / 3.2.3: |
Intelligent K-Means / 3.3: |
Iterated Anomalous pattern for iK-Means / 3.3.1: |
Cross validation of iK-Means results / 3.3.2: |
Interpretation aids / 3.4: |
Conventional interpretation aids / 3.4.1: |
Contribution and relative contribution tables / 3.4.2: |
Cluster representatives / 3.4.3: |
Measures of association from ScaD tables / 3.4.4: |
Overall assessment / 3.5: |
Ward Hierarchical Clustering / 4: |
Agglomeration: Ward algorithm / 4.1: |
Divisive clustering with Ward criterion / 4.2: |
2-Means splitting / 4.2.1: |
Splitting by separating / 4.2.2: |
Interpretation aids for upper cluster hierarchies / 4.2.3: |
Conceptual clustering / 4.3: |
Extensions of Ward clustering / 4.4: |
Agglomerative clustering with dissimilarity data / 4.4.1: |
Hierarchical clustering for contingency and flow data / 4.4.2: |
Data Recovery Models / 4.5: |
Statistics modeling as data recovery / 5.1: |
Averaging / 5.1.1: |
Linear regression / 5.1.2: |
Principal component analysis / 5.1.3: |
Correspondence factor analysis / 5.1.4: |
Data recovery model for K-Means / 5.2: |
Equation and data scatter decomposition / 5.2.1: |
Contributions of clusters, features, and individual entities / 5.2.2: |
Correlation ratio as contribution / 5.2.3: |
Partition contingency coefficients / 5.2.4: |
Data recovery models for Ward criterion / 5.3: |
Data recovery models with cluster hierarchies / 5.3.1: |
Covariances, variances and data scatter decomposed / 5.3.2: |
Direct proof of the equivalence between 2-Means and Ward criteria / 5.3.3: |
Gower's controversy / 5.3.4: |
Extensions to other data types / 5.4: |
Similarity and attraction measures compatible with K-Means and Ward criteria / 5.4.1: |
Application to binary data / 5.4.2: |
Agglomeration and aggregation of contingency data / 5.4.3: |
Extension to multiple data / 5.4.4: |
One-by-one clustering / 5.5: |
PCA and data recovery clustering / 5.5.1: |
Divisive Ward-like clustering / 5.5.2: |
Iterated Anomalous pattern / 5.5.3: |
Anomalous pattern versus Splitting / 5.5.4: |
One-by-one clusters for similarity data / 5.5.5: |
Different Clustering Approaches / 5.6: |
Extensions of K-Means clustering / 6.1: |
Clustering criteria and implementation / 6.1.1: |
Partitioning around medoids PAM / 6.1.2: |
Fuzzy clustering / 6.1.3: |
Regression-wise clustering / 6.1.4: |
Mixture of distributions and EM algorithm / 6.1.5: |
Kohonen self-organizing maps SOM / 6.1.6: |
Graph-theoretic approaches / 6.2: |
Single linkage, minimum spanning tree and connected components / 6.2.1: |
Finding a core / 6.2.2: |
Conceptual description of clusters / 6.3: |
False positives and negatives / 6.3.1: |
Conceptually describing a partition / 6.3.2: |
Describing a cluster with production rules / 6.3.3: |
Comprehensive conjunctive description of a cluster / 6.3.4: |
General Issues / 6.4: |
Feature selection and extraction / 7.1: |
A review / 7.1.1: |
Comprehensive description as a feature selector / 7.1.2: |
Comprehensive description as a feature extractor / 7.1.3: |
Data pre-processing and standardization / 7.2: |
Dis/similarity between entities / 7.2.1: |
Pre-processing feature based data / 7.2.2: |
Data standardization / 7.2.3: |
Similarity on subsets and partitions / 7.3: |
Dis/similarity between binary entities or subsets / 7.3.1: |
Dis/similarity between partitions / 7.3.2: |
Dealing with missing data / 7.4: |
Imputation as part of pre-processing / 7.4.1: |
Conditional mean / 7.4.2: |
Maximum likelihood / 7.4.3: |
Least-squares approximation / 7.4.4: |
Validity and reliability / 7.5: |
Index based validation / 7.5.1: |
Resampling for validation and selection / 7.5.2: |
Model selection with resampling / 7.5.3: |
Conclusion: Data Recovery Approach in Clustering / 7.6: |
Bibliography |
Index |
Preface |
List of Denotations |
Introduction: Historical Remarks |
What Is Clustering / 1: |
Base words |
Exemplary problems / 1.1: |