This book constitutes the refereed proceedings of the 18th European Conference on Machine Learning, ECML 2007, held in Warsaw, Poland, September 17-21, 2007, jointly with PKDD 2007. The 41 revised full papers and 37 revised short papers presented together with abstracts of 4 invited talks were carefully reviewed and selected from 592 abstracts submitted to both, ECML and PKDD. The papers present a wealth of new results in the area and address all current issues in machine learning. The two premier annual European conferences in the areas of machine learning and data mining have been collocated ever since the ?rst joint conference in Freiburg, 2001. The European Conference on Machine Learning (ECML) traces its origins to 1986, when the ?rst European Working Session on Learning was held in Orsay, France. The European Conference on Principles and Practice of KnowledgeDiscoveryinDatabases(PKDD) was?rstheldin1997inTrondheim, Norway. Over the years, the ECML/PKDD series has evolved into one of the largest and most selective international conferences in machine learning and data mining. In 2007, the seventh collocated ECML/PKDD took place during September 17¿21 on the centralcampus of WarsawUniversityand in the nearby Staszic Palace of the Polish Academy of Sciences. The conference for the third time used a hierarchical reviewing process. We nominated 30 Area Chairs, each of them responsible for one sub-?eld or several closely related research topics. Suitable areas were selected on the basis of the submission statistics for ECML/PKDD 2006 and for last year¿s International Conference on Machine Learning (ICML 2006) to ensure a proper load balance amongtheAreaChairs.AjointProgramCommittee(PC)wasnominatedforthe two conferences, consisting of some 300 renowned researchers, mostly proposed by the Area Chairs. This joint PC, the largest of the series to date, allowed us to exploit synergies and deal competently with topic overlaps between ECML and PKDD. ECML/PKDD 2007 received 592 abstract submissions. As in previous years, toassistthereviewersandtheAreaChairsintheir?nalrecommendationauthors had the opportunity to communicate their feedback after the reviewing phase. Invited Talks Learning, Information Extraction and the Web 1 Tom M. Mitchell Putting Things in Order: On the Fundamental Role of Ranking in Classification and Probability Estimation 2 Peter A. Flach Mining Queries 4 Ricardo Baeza-Yates Adventures in Personalized Information Access 5 Barry Smyth Long Papers Statistical Debugging Using Latent Topic Models 6 David Andrzejewski, Anne Mulhern, Ben Liblit, and Xiaojin Zhu Learning Balls of Strings with Correction Qiieries 18 Leonor Becerra Bonache, Colin, de la Higuera, Jean-Christophe Janodet, and Fr ric Tantini Neighborhood-Based Local Sensitivity 30 Paul N. Bennett Approximating Gaussian Processes with H -Matricesr 42 Steffen B rn and Jochen Garcke Learning Metrics Between Tree Structured Data: Application to Image Recognition 54 Laurent Boyer, Amaury Habrard, and Mare Sebban Shrinkage Estimator for Bayesian Network Parameters 67 John Burge, Terran Lane Level Learning Set: A Novel Classifier Based on Active Contour Models 79 Xiongcai Cai and Arcot Sowmya Learning Partially Observable Markov Models from First Passage Times 91 J me Callut and Pierre Dupont Context Sensitive Paraphrasing with a Global Unsupervised Classifier 104 Michael Connor and Dan Roth Dual Strategy Active Learning 116 Pinar Donmez, Jaime G. Carbonell, and Paul N. Bennett Decision Tree Instability and Active Learning 128 Kenneth Dwyer and Robert Holte Constraint Selection by Committee: An Ensemble Approach to Identifying Informative Constraints for Semi-supervised Clustering 140 Derek Greene and P aig Cunningham The Cost of Learning Directed Cuts 152 Thomas G ner and Gemma C. Garriga Spectral Clustering and Embedding with Hidden Markov Models 164 Tony Jebara, Yingbo Song, and Kapil Thadani Probabilistic Explanation Based Learning 176 Angelika Kimmig, Luc De Raedt, and Hannu Toivonen Graph-Based Domain Mapping for Transfer Learning in General Games 188 Gregory Kuhlmann and Peter Stone Learning to Classify Documents with Only a Small Positive Training Set 201 Xiao-Li Li, Bing Liu, and See-Kiong Ng Structure Learning of Probabilistic Relational Models from Incomplete Relational Data 214 Xiao-Lin Li and Zhi-Hua Zhou Stability Based Sparse LSI/PCA: Incorporating Feature Selection in LSI and PCA 226 Dimitrios Mavroeidis and Michalis Vazirgiannis Bayesian Substructure Learning - Approximate Learning of Very Large Network Structures 238 Andreas N le, Math Dejori, and Martin Stetter Efficient Continuous-Time Reinforcement Learning with Adaptive State Graphs 250 Gerhard Neumann, Michael Pfeiffer, and Wolfgang Maass Source Separation with Gaussian Process Models 262 Sunho Park and Seungjin Choi Discriminative Sequence Labeling by Z-Score Optimization 274 Elisa Ricci, Tijl de Bie, and Nello Cristianini Fast Optimization Methods for L1 Regularization: A Comparatie Study and Two New Approaches 286 Mark Schmidt, Glenn Fung, and R mer Rosales Bayesian Inference for Sparse Generalized Linear Models 298 Matthias Seeger, Sebastian Gerwinn, and Matthias Bethge Classifier Loss Under Metric Uncertainty 310 David B. Skalak, Alexandru Niculescu-Mizil, and Rich Caruana Additive Groves of Regression Trees 323 Darla Sorokina, Rich Caruana, and Mirek Riedewald Efficient Computation of Recursive Principal Component Analysis for Structured Input 335 Alessandro Sperduti Hinge Rank Loss and the Area Under the ROC Curve 347 Harald Steck Clustering Trees with Instance Level Constraints 359 Jan Struyf and Sa o D eroski On Pairwise Naive Bayes Classifiers 371 Jan-Nikolas Sulzmann, Johannes F rnkranz, and Eyke H llermeier Separating Precision and Mean in Dirichlet-Enhanced High-Order Markov Models 382 Rikiya Takahashi Safe Q-Learning on Complete History Spaces 394 Stephan Timmer and Martin Riedmiller Random k-Labelsets: An Ensemble Method for Multilabel Classification 406 Grigorios Tsoumakas and Ioannis Vlahavas Seeing the Forest Through the Trees: Learning a Comprehensible Model from an Ensemble 418 Anneleen Van Assche and Hendrik Blockeel Avoiding Boosting Overfitting by Removing Confusing Samples 430 Alexander Vezhnevets and Olga Barinova Planning and Learning in Environments with Delayed Feedback 442 Thomas J. Walsh, Ali Nouri, Lihong Li, and Michael L. Littman Analysing Co-training Style Algorithms 454 Wei Wang and Zhi-Hua Zhou Policy Gradient Critics 466 Own Wierstra and J rgen Schmidhuber An Improved Model Selection Heuristic for AUC 478 Shaomin Wu, Peter Flach, and C r Ferri Finding the Right Family: Parent and Child Selection for Averaged One-Dependence Estimators 490 Fei Zheng and Geoffrey I. Webb Short Papers Stepwise Induction of Multi-target Model Trees 502 Annalisa Appice and Saso D eroski Comparing Rule Measures for Predictive Association Rules 510 Paulo J. Azevedo and Alipio M. Jorge User Oriented Hierarchical Information Organization and Retrieval 518 Korinna Bade, Marcel Hermkes, and Andreas N rnberger Learning a Classifier with Very Few Examples: Analogy Based and Knowledge Based Generation of New Examples for Character Recognition 527 S. Bayoudh, H. Mouch , L. Miclet, and E. Anquetil Weighted Kernel Regression for Predicting Changing Dependencies 535 Steven Busuttil and Yuri Kalnishkan Counter-Example Generation-Based One-Class Classification 543 Andr B almi, Andr Kocsor, and R bert Busa-Fekete Test-Cost Sensitive Classification Based on Conditioned Loss Functions 551 Mumin Cebe and Cigdem, Gunduz-Demir Probabilistic Models for Action-Based Chinese Dependency Parsing 559 Xiangyu Duan, Jun Zhao, and Bo Xu Learning Directed Probabilistic Logical Models: Ordering-Search Versus Structure-Search 567 Dawn, Fierens, Jan Ramon, Maurice Bruynooghe, and Hendrik Blockeel A Simple Lexicographic Ranker and Probability Estimator 575 Peter Flach and Edson Takashi Matsubara On Minimizing the Position Error in Label Ranking 583 Eyke H llermeier and Johannes F rnkranz On Phase Transitions in Learning Sparse Networks 591 Goele Hollanders, Geert Jan Bex, Marc Gyssens, Ronald L. Westra, and Karl Tuyls Semi-supervised Collaborative Text Classification 600 Rong Jin, Ming Wu, and Rahul Sukthankar Learning from Relevant Tasks Only 608 Samuel Kaski and Jaakko Peltonen An Unsupervised Learning Algorithm for Rank Aggregation 616 Alexandre Klementiev, Dan Roth, and Kevin Small Ensembles of Multi-Objective Decision Trees 624 Dragi Kocev, Celine Vens, Jan Struyf, and Sa o D eroski Kernel-Based Grouping of Histogram Data 632 Tilman Lange and Joachim M. Buhmann Active Class Selection 640 R. Lomasky, C.E. Brodley, M. Aernecke, D. Walt, and M. Friedl Sequence Labeling with Reinforcement Learning and Ranking Algorithms 648 Francis Maes, Ludovic Denoyer, and Patrick Gallinari Efficient Pairwise Classification 658 Sang-Hyeun Park and Johannes F rnkranz Scale-Space Based Weak Regressors for Boosting 666 Jin-Hyeong Park and Chandan K. Reddy K-Means with Large and Noisy Constraint Sets 674 Dan Pelleg and Dorit Baras Towards 'Interactive' Active Learning in Multi-view Feature Sets for Information Extraction 683 Katharine Probst and Rayid Ghani Principal Component Analysis for Large Scale Problems with Lots of Missing Values 691 Tapani Raiko, Alexander Ilin, and Juha Karhunen Transfer Learning in Reinforcement Learning Problems Through Partial Policy Recycling 699 Jan Ramon, Kurt Driessens, and Tom Croonenborghs Class Noise Mitigation Through Instance Weighting 708 Umaa Rebbapragada, and Carla E. Brodley Optimizing Feature Sets for Structured Data 716 Ulrich R ckert and Stefan Kramer Roulette Sampling for Cost-Sensitive Learning 724 Victor S. Sheng and Charles X. Ling Modeling Highway Traffic Volumes 732 Tom ingliar and Milo Hauskrecht Undercomplete Blind Subspace Deconvolution Via Linear Prediction 740 Zolt Szab , Barnab P czos, and Andr L rincz Learning an Outlier-Robust Kalman Filter 748 Jo-Anne Ting, Evangelos Theodorou, and Stefan Schaal Imitation Learning Using Graphical Models 757 Deepak Verma and Rajesh P.N. Rao Nondeterministic Discretization of Weights Improves Accuracy of Neural Networks 765 Marcin Wojnarski Semi-definite Manifold Alignment 773 Liang Xiong, Fei Wang, and Changshui Zhang General Solution for Supervised Graph Embedding 782 Qubo You, Nanning Zheng, Shaoyi Du. and Yang Wu Multi-objective Genetic Programming for Multiple Instance Learning 790 Amelia Zufra, and Sebasti Ventura Exploiting Term, Predicate, and Feature Taxonomies in Propositionalization and Propositional Rule Learning 798 Monika akov nd Filip elezn Author Index 807