Inductive logic programming

The book features a collection of invited papers and research contributions focused on various aspects of Inductive Logic Programming (ILP). It discusses personal insights on effectively applying ILP, along with topics such as reasoning and learning agents. Key research papers explore multi-relational approaches to mining model trees, complexity parameters for first-order classes, and the implementation of a multi-relational decision tree learning algorithm. The application of theory revision to distributed database design is examined, alongside disjunctive learning using a soft-clustering method. Other areas of interest include mathematical discovery through ILP, an exhaustive matching procedure to enhance learning efficiency, and efficient data structures for ILP. The book also covers graph kernels and Gaussian processes for relational reinforcement learning, clause condensation, and a comparative evaluation of feature set evolution strategies for multirelational boosting. Additional topics include approaches to propositionalization, ideal refinement of descriptions in -Log, and learning first-order logic clauses via genetic algorithms. The induction of enzyme classes from biological databases, estimation of maximum likelihood parameters for stochastic context-free graph grammars, and hybrid abductive inductive learning are also discussed. Lastly, it addresses query optimization in ILP and efficient learning of unlabeled ter