Machine Learning: ECML 2004 Book

	Random matrices in data analysis	1
	Data privacy	8
	Breaking through the syntax barrier : searching with entities and relations	9
	Real-world learning with Markov logic networks	17
	Strength in diversity : the advance of data analysis	18
	Filtered reinforcement learning	27
	Applying support vector machines to imbalanced datasets	39
	Sensitivity analysis of the result in binary decision trees	51
	A boosting approach to multiple instance learning	63
	An experimental study of different approaches to reinforcement learning in common interest stochastic games	75
	Learning from message pairs for automatic email answering	87
	Concept formation in expressive description logics	99
	Multi-level boundary classification for information extraction	111
	An analysis of stopping and filtering criteria for rule learning	123
	Adaptive online time allocation to search algorithms	134
	Model approximation for HEXQ hierarchical reinforcement learning	144
	Iterative ensemble classification for relational data : a case study of semantic Web services	156
	Analyzing multi-agent reinforcement learning using evolutionary dynamics	168
	Experiments in value function approximation with sparse support vector regression	180
	Constructive induction for classifying time series	192
	Fisher kernels for logical sequences	205
	The Enron corpus : a new dataset for email classification research	217
	Margin maximizing discriminant analysis	227
	Multi-objective classification with info-fuzzy networks	239
	Improving progressive sampling via meta-learning on learning curves	250
	Methods for rule conflict resolution	262
	An efficient method to estimate labelled sample size for transductive LDA(QDA/MDA) : based on Bayes risk	274
	Analyzing sensory data using non-linear preference learning with feature subset selection	286
	Dynamic asset allocation exploiting predictors in reinforcement learning framework	298
	Justification-based selection of training examples for case base reduction	310
	Using feature conjunctions across examples for learning pairwise classifiers	322
	Feature selection filters based on the permutation test	334
	Sparse distributed memories for on-line value-based reinforcement learning	347
	Improving random forests	359
	The principal components analysis of a graph, and its relationships to spectral clustering	371
	Using string kernels to identify famous performers from their playing style	384
	Associative clustering	396
	Learning to fly simple and robust	407
	Bayesian network methods for traffic flow forecasting with incomplete data	419
	Matching model versus single model : a study of the requirement to match class distribution using decision trees	429
	Inducing polynomial equations for regression	441
	Efficient hyperkernel learning using second-order cone programming	453
	Effective voting of heterogeneous classifiers	465
	Convergence and divergence in standard and averaging reinforcement learning	477
	Document representation for one-class SVM	489
	Naive Bayesian classifiers for ranking	501
	Conditional independence trees	513
	Exploiting unlabeled data in content-based image retrieval	525
	Population diversity in permutation-based genetic algorithm	537
	Simultaneous concept learning of fuzzy rules	548
	SWITCH : a novel approach to ensemble learning for heterogeneous data	560
	Estimating attributed central orders - an empirical comparison	563
	Batch reinforcement learning with state importance	566
	Explicit local models : towards "optimal" optimization algorithms	569
	An intelligent model for the Signorini contact problem in belt grinding processes	572
	Cluster-grouping : from subgroup discovery to clustering	575