Machine Learning and Natural Language Processing Lab

Relational Cond... |--| Asynchrone Rela... |--| Flexible String... |--| Lernen von Inte... |--| Speeding Up Log... |--| Relational Fish... |--| Development of ... |--| Development of ... |--| Collective Clas... |--| Learning Locali... |--| Using Equation ... |--| Classification ... |--| Induktive Daten... |--| Constraint-Base... |--| Magic Bayes-Bal... |--| Logistic Model ... |--| Clustering unte... |--| Two-Level Class... |--| Machine Learnin... |--| Erweiterung des... |--| Bayes'sche-logi... |--|

Master Thesis

Two-Level Classification for Generalized Multi-Instance Data

Nils B. Weidmann, 2003

In standard multi-instance (MI) learning, a single positive instance in a bag of instances produces a positive class label for the bag. Hence, the learner knows how the bag s class label is determined by the (unobservable) labels of the instances and can explicitly use this information to solve the learning task. In this thesis, a generalized view of the MI problem is investigated, where this no longer holds, and an interaction between instances in a bag is assumed to determine the bag label. A two-level learning method (TLC) is presented. TLC transforms a bag into a single meta-instance that can then be passed on to a propositional learning method. The meta-instance contains information about the number of instances in the bag that belong to certain regions in the instance space. Experiments with TLC were performed on a variety of datasets. TLC achieves competitive results, and outperforms all existing approaches on the small Mutagenesis dataset. References: Dietterich, T. G., Lathrop, R. H. and Lozano-Perez, T. [1997]. Solving the Multiple Instance Problem with Axis-Parallel Rectangles. Artificial Intelligence, 89(1-2), 31-71.