Distributed clustering algorithms.

by Chan Wai To.Thesis (M.Phil.)--Chinese University of Hong Kong, 2001.Includes bibliographical references (leaves 117-121).Abstracts in English and Chinese.Abstract --- p.iiAcknowledgments --- p.ivChapter 1 --- Introduction --- p.1Chapter 1.1 --- Clustering --- p.3Chapter 1.2 --- Mobile Agent --- p.4Chapter 1.3 --- Contribution --- p.4Chapter 1.4 --- Outline of this Thesis --- p.5Chapter 2 --- Related Work --- p.6Chapter 2.1 --- Clustering --- p.6Chapter 2.1.1 --- K-Means Clustering --- p.6Chapter 2.1.2 --- A more efficient K-Means Clustering Algorithm --- p.3Chapter 2.1.3 --- K-Medoids Clustering Algorithms --- p.8Chapter 2.1.4 --- Linkage-based Methods --- p.11Chapter 2.1.5 --- BIRCH --- p.13Chapter 2.1.6 --- DBSCAN --- p.14Chapter 2.1.7 --- Other Clustering Algorithm --- p.17Chapter 2.2 --- Parallel Clustering and Distributed Clustering --- p.17Chapter 2.2.1 --- A Fast Parallel Clustering Algorithm for Large Spatial Databases --- p.17Chapter 2.3 --- Distributed Data Mining --- p.18Chapter 2.3.1 --- A Distributed Clustering Algorithm --- p.18Chapter 2.3.2 --- Efficient Mining of Association Rules in Distributed Databases --- p.19Chapter 2.4 --- Information Retrieval and Document Clustering --- p.20Chapter 2.4.1 --- Document and Document Set Representation --- p.20Chapter 2.4.2 --- TFIDF --- p.20Chapter 2.4.3 --- Similarity --- p.21Chapter 2.4.4 --- Partitional Document Clustering --- p.22Chapter 2.4.5 --- Hierarchical Document Clustering --- p.22Chapter 2.4.6 --- Document Clustering Application --- p.23Chapter 3 --- Distributed Clustering --- p.24Chapter 3.1 --- Problem Description --- p.24Chapter 3.2 --- Distributed k-Means Clustering Algorithm --- p.25Chapter 3.2.1 --- Initialization --- p.25Chapter 3.2.2 --- weighted k-Means procedure --- p.26Chapter 3.2.3 --- Refinement --- p.27Chapter 3.2.4 --- Example --- p.31Chapter 3.2.5 --- Communication Cost --- p.34Chapter 3.3 --- Grid k-Mean --- p.34Chapter 3.3.1 --- Runtime Splitting --- p.36Chapter 3.3.2 --- Initial Clusters --- p.38Chapter 3.3.3 --- Refinement --- p.38Chapter 3.3.4 --- Overall Algorithm --- p.39Chapter 3.3.5 --- Efficiency in Decomposition --- p.42Chapter 3.3.6 --- Example --- p.42Chapter 3.3.7 --- Comparison with previous k-Means method --- p.43Chapter 3.3.8 --- Communication Cost --- p.44Chapter 3.4 --- Experiment --- p.44Chapter 3.4.1 --- Performance --- p.46Chapter 3.4.2 --- Communication Cost --- p.47Chapter 3.4.3 --- Quality of Clustering --- p.49Chapter 3.4.4 --- Clustering in High Dimension --- p.49Chapter 3.4.5 --- Other Data Distributions --- p.52Chapter 4 --- Distributed DBSCAN --- p.54Chapter 4.1 --- Representative points of local candidate clusters --- p.55Chapter 4.2 --- Verification and Cluster Merging --- p.57Chapter 4.2.1 --- Clustering Result Quality --- p.59Chapter 4.3 --- Experiment --- p.62Chapter 5 --- Document Clustering --- p.72Chapter 5.1 --- Initialization --- p.73Chapter 5.2 --- Refinement --- p.76Chapter 5.3 --- Stopping criteria --- p.77Chapter 5.4 --- Message --- p.77Chapter 5.5 --- Algorithm --- p.78Chapter 5.6 --- Experiment --- p.82Chapter 5.6.1 --- Data Source and Experimental Setup --- p.82Chapter 5.6.2 --- Data Size --- p.34Chapter 5.6.3 --- Evaluation Metrics --- p.85Chapter 5.6.4 --- Experimental Result --- p.85Chapter 5.6.5 --- Comparison to Other Algorithms --- p.94Chapter 5.6.6 --- Conclusion --- p.94Chapter 5.7 --- Future Work --- p.95Chapter 6 --- Agent and Implementation Details --- p.96Chapter 6.1 --- Agent Introduction --- p.96Chapter 6.1.1 --- Reason to use Mobile Agent --- p.97Chapter 6.1.2 --- Grasshopper Overview --- p.97Chapter 6.1.3 --- Agent Scenario --- p.98Chapter 6.1.4 --- Another Agent Scenario --- p.99Chapter 6.2 --- Implementation Details --- p.100Chapter 6.2.1 --- Distributed k-Means --- p.100Chapter 6.2.2 --- Grid k-Means --- p.104Chapter 6.2.3 --- Distributed DBSCAN --- p.109Chapter 6.2.4 --- Distributed Document Clustering --- p.112Chapter 7 --- Conclusio