Previous edition: New York: Wiley, 2002.

Includes bibliographical references and index.

Print version record.

Note continued: 9.3.Comparing Classifier Performance -- 9.3.1. Which Technique is Best? -- 9.3.2. Statistical Tests -- 9.3.3.Comparing Rules When Misclassification Costs are Uncertain -- 9.3.4. Example Application Study -- 9.3.5. Further Developments -- 9.3.6. Summary -- 9.4. Application Studies -- 9.5. Summary and Discussion -- 9.6. Recommendations -- 9.7. Notes and References -- Exercises -- 10. Feature Selection and Extraction -- 10.1. Introduction -- 10.2. Feature Selection -- 10.2.1. Introduction -- 10.2.2. Characterisation of Feature Selection Approaches -- 10.2.3. Evaluation Measures -- 10.2.4. Search Algorithms for Feature Subset Selection -- 10.2.5.Complete Search -- Branch and Bound -- 10.2.6. Sequential Search -- 10.2.7. Random Search -- 10.2.8. Markov Blanket -- 10.2.9. Stability of Feature Selection -- 10.2.10. Example Application Study -- 10.2.11. Further Developments -- 10.2.12. Summary -- 10.3. Linear Feature Extraction -- 10.3.1. Principal Components Analysis -- 10.3.2. Karhunen-Loeve Transformation -- 10.3.3. Example Application Study -- 10.3.4. Further Developments -- 10.3.5. Summary -- 10.4. Multidimensional Scaling -- 10.4.1. Classical Scaling -- 10.4.2. Metric MDS -- 10.4.3. Ordinal Scaling -- 10.4.4. Algorithms -- 10.4.5. MDS for Feature Extraction -- 10.4.6. Example Application Study -- 10.4.7. Further Developments -- 10.4.8. Summary -- 10.5. Application Studies -- 10.6. Summary and Discussion -- 10.7. Recommendations -- 10.8. Notes and References -- Exercises -- 11. Clustering -- 11.1. Introduction -- 11.2. Hierarchical Methods -- 11.2.1. Single-Link Method -- 11.2.2.Complete-Link Method -- 11.2.3. Sum-of-Squares Method -- 11.2.4. General Agglomerative Algorithm -- 11.2.5. Properties of a Hierarchical Classification -- 11.2.6. Example Application Study -- 11.2.7. Summary -- 11.3. Quick Partitions -- 11.4. Mixture Models -- 11.4.1. Model Description -- 11.4.2. Example Application Study -- 11.5. Sum-of-Squares Methods -- 11.5.1. Clustering Criteria -- 11.5.2. Clustering Algorithms -- 11.5.3. Vector Quantisation -- 11.5.4. Example Application Study -- 11.5.5. Further Developments -- 11.5.6. Summary -- 11.6. Spectral Clustering -- 11.6.1. Elementary Graph Theory -- 11.6.2. Similarity Matrices -- 11.6.3. Application to Clustering -- 11.6.4. Spectral Clustering Algorithm -- 11.6.5. Forms of Graph Laplacian -- 11.6.6. Example Application Study -- 11.6.7. Further Developments -- 11.6.8. Summary -- 11.7. Cluster Validity -- 11.7.1. Introduction -- 11.7.2. Statistical Tests -- 11.7.3. Absence of Class Structure -- 11.7.4. Validity of Individual Clusters -- 11.7.5. Hierarchical Clustering -- 11.7.6. Validation of Individual Clusterings -- 11.7.7. Partitions -- 11.7.8. Relative Criteria -- 11.7.9. Choosing the Number of Clusters -- 11.8. Application Studies -- 11.9. Summary and Discussion -- 11.10. Recommendations -- 11.11. Notes and References -- Exercises -- 12.Complex Networks -- 12.1. Introduction -- 12.1.1. Characteristics -- 12.1.2. Properties -- 12.1.3. Questions to Address -- 12.1.4. Descriptive Features -- 12.1.5. Outline -- 12.2. Mathematics of Networks -- 12.2.1. Graph Matrices -- 12.2.2. Connectivity -- 12.2.3. Distance Measures -- 12.2.4. Weighted Networks -- 12.2.5. Centrality Measures -- 12.2.6. Random Graphs -- 12.3.Community Detection -- 12.3.1. Clustering Methods -- 12.3.2. Girvan-Newman Algorithm -- 12.3.3. Modularity Approaches -- 12.3.4. Local Modularity -- 12.3.5. Clique Percolation -- 12.3.6. Example Application Study -- 12.3.7. Further Developments -- 12.3.8. Summary -- 12.4. Link Prediction -- 12.4.1. Approaches to Link Prediction -- 12.4.2. Example Application Study -- 12.4.3. Further Developments -- 12.5. Application Studies -- 12.6. Summary and Discussion -- 12.7. Recommendations -- 12.8. Notes and References -- Exercises -- 13. Additional Topics -- 13.1. Model Selection -- 13.1.1. Separate Training and Test Sets -- 13.1.2. Cross-Validation -- 13.1.3. The Bayesian Viewpoint -- 13.1.4. Akaike's Information Criterion -- 13.1.5. Minimum Description Length -- 13.2. Missing Data -- 13.3. Outlier Detection and Robust Procedures -- 13.4. Mixed Continuous and Discrete Variables -- 13.5. Structural Risk Minimisation and the Vapnik-Chervonenkis Dimension -- 13.5.1. Bounds on the Expected Risk -- 13.5.2. The VC Dimension.

"Statistical Pattern Recognition provides an introduction to statistical pattern theory and techniques, with material drawn from a wide range of fields, including the areas of engineering, statistics, computer science and the social sciences. The book describes techniques for analysing data comprising measurements made on individuals or objects. The techniques are used to make a prediction such as disease of a patient, the type of object illuminated by a radar, economic forecast. Emphasis is placed on techniques for classification, a term used for predicting the class or group an object belongs to (based on a set of exemplars) and for methods that seek to discover natural groupings in a data set. Each section concludes with a description of the wide range of practical applications that have been addressed and the further developments of theoretical techniques and includes a variety of exercises, from 'open-book' questions to more lengthy projects. New material is presented, including the analysis of complex networks and basic techniques for analysing the properties of datasets and also introduces readers to the use of variational methods for Bayesian density estimation and looks at new applications in biometrics and security."-- Provided by publisher.

"The book describes techniques for analysing data comprising measurements made on individuals or objects."-- Provided by publisher.

Social Sciences and Humanities