Constrained Clustering: Advances in Algorithms, Theory, and ApplicationsSugato Basu, Ian Davidson, Kiri Wagstaff CRC Press, 18. aug. 2008 - 472 sider This volume encompasses many new types of constraints and clustering methods as well as delivers thorough coverage of the capabilities and limitations of constrained clustering. With contributions from industrial researchers and leading academic experts who pioneered the field, it provides a well-balanced combination of theoretical advances, key algorithmic development, and novel applications. The book presents various types of constraints for clustering and describes useful variations of the standard problem of clustering under constraints. It also demonstrates the application of clustering with constraints to relational, bibliographic, and video data. |
Indhold
Introduction | 1 |
SemiSupervised Clustering with User Feedback | 17 |
Gaussian Mixture Models with Equivalence Constraints | 33 |
Pairwise Constraints as Priors in Probabilistic Clustering | 59 |
Clustering with Constraints A MeanField Approximation Perspective | 91 |
ConstraintDriven CoClustering of 01 Data | 123 |
On Supervised Clustering for Creating Categorization Segmentations | 149 |
Clustering with Balancing Constraints | 171 |
Collective Relational Clustering | 221 |
NonRedundant Data Clustering | 245 |
Joint Cluster Analysis of Attribute Data and Relationship Data | 285 |
Correlation Clustering | 313 |
Interactive Visual Clustering for Relational Data | 329 |
Distance Metric Learning from CannotbeLinked Example Pairs with Application to Name Disambiguation | 357 |
PrivacyPreserving Data Publishing A ConstraintBased Clustering Approach | 375 |
Learning with Pairwise Constraints for Video Object Classification | 397 |
Andre udgaver - Se alle
Constrained Clustering: Advances in Algorithms, Theory, and Applications Sugato Basu,Ian Davidson,Kiri Wagstaff Ingen forhåndsvisning - 2008 |
Almindelige termer og sætninger
approach approximation attributes balanced clustering bi-sets cannot-link constraints CCIB chunklet classification clus cluster assignment cluster centers clustering algorithm clustering methods clustering problem co-clustering Computer Science Conference on Machine constrained clustering constrained k-means Correlation Clustering data clustering Data Mining data points data set database denote distance metric documents edges EM algorithm entity resolution equation equivalence constraints example Figure Gaussian Gibbs sampling graph IEEE instances International Conference iteration k-means algorithm Knowledge Discovery labeled data Layout learning algorithms loss function Machine Learning Markov network matrix metric learning minimize mixture model must-link constraints mutual information nodes number of clusters objective function optimization pairs pairwise constraints pairwise loss functions pairwise relations parameter partition performance probabilistic Proceedings Rand index references sampling Section semi-supervised clustering semi-supervised learning similar solution straints supervised clustering supervised learning Ti,h update values vector weight Yahoo μη