discovery. The goal of subgroup discovery is to find rules describing subsets of the population that are sufficiently large and statistically unusual. The paper presents a subgroup discovery algorithm, CN2-SD, developed by modifying parts of the CN2 classification rule learner: its covering algorithm, search heuristic, probabilistic classification of instances, and evaluation measures. Experimental evaluation of CN2-SD on 23 UCI data sets shows substantial reduction of the number of induced rules, increased rule coverage and rule significance, as well as slight improvements in terms of the area under ROC curve, when compared with the CN2 algorithm. Application of CN2-SD to a large traffic accident data set confirms these findings.

Propositionalization has already been shown to be a promising approach for robustly and e#ectively handling relational data sets for knowledge discovery. In this paper, we compare up-to-date methods for propositionalization from two main groups: logic-oriented and databaseoriented techniques. Experiments using several learning tasks --- both ILP benchmarks and tasks from recent international data mining competitions --- show that both groups have their specific advantages. While logic-oriented methods can handle complex background knowledge and provide expressive first-order models, database-oriented methods can be more e#cient especially on larger data sets. Obtained accuracies vary such that a combination of the features produced by both groups seems a further valuable venture.

Many domains in the field of Inductive Logic Programming (ILP) involve highly unbalanced data. Our research has focused on Information Extraction (IE), a task that typically involves many more negative examples than positive examples. IE is the process of finding facts in unstructured text, such as biomedical journals, and putting those facts in an organized system. In particular, we have focused on learning to recognize instances of the protein-localization relationship in Medline abstracts. We view the problem as a machine-learning task: given positive and negative extractions from a training corpus of abstracts, learn a logical theory that performs well on a held-aside testing set. A common way to measure performance in these domains is to use precision and recall instead of simply using accuracy. We propose Gleaner, a randomized search method which collects good clauses from a broad spectrum of points along the recall dimension in recall-precision curves and employs an "at least N of these M clauses" thresholding method to combine the selected clauses. We compare Gleaner to ensembles of standard Aleph theories and find that Gleaner produces comparable testset results in a fraction of the training time needed for ensembles.

In this paper we present 1BC and 1BC2, two systems that perform naive Bayesian classification of structured individuals. The approach of 1BC is to project the individuals along first-order features. These features are built from the individual using structural predicates referring to related objects (e.g. atoms within molecules), and properties applying to the individual or one or several of its related objects (e.g. a bond between two atoms). We describe an individual in terms of elementary features consisting of zero or more structural predicates and one property; these features are treated as conditionally independent in the spirit of the naive Bayes assumption. 1BC2 represents an alternative first-order upgrade to the naive Bayesian classifier by considering probability distributions over structured objects (e.g., a molecule as a set of atoms), and estimating those distributions from the probabilities of its elements (which are assumed to be independent). We present a unifying view on both systems in which 1BC works in language space, and 1BC2 works in individual space. We also present a new, efficient recursive algorithm improving upon the original propositionalisation approach of 1BC. Both systems have been implemented in the context of the first-order descriptive learner Tertius, and we investigate the differences between the two systems both in computational terms and on artificially generated data. Finally, we describe a range of experiments on ILP benchmark data sets demonstrating the viability of our approach.

Subgroup discovery aims at finding interesting subsets of a classified example set that deviates from the overall distribution. The search is guided by a so-called utility function, trading the size of subsets (coverage) against their statistical unusualness. By choosing the utility function accordingly, subgroup discovery is well suited to find interesting rules with much smaller coverage and bias than possible with standard classifier induction algorithms. Smaller subsets can be considered local patterns, but this work uses yet another definition: According to this definition global patterns consist of all patterns reflecting the prior knowledge available to a learner, including all previously found patterns.

ILP nepraktickymi. Algoritmy ILP naprklad casto vyzaduj, aby uzivatel reprezentacn jazyk vymezil komplikovanymi deklaracemi. Prizpusobit existujc nastroje pozadavkum uzivatelu na jednoduche ovladan je (mozna prekvapive) tezkym ukolem, o cemz svedc napr. nedavny neuspech projektu zamereneho na vclenen znameho ILP programu Progol [11] do softwaroveho balku Clementine [3]. Setka-li se dobyvatel znalost s ulohou vyhledat souvislosti rozprostrene v nekolika relacch datab aze, v typickem prpade potrebne relace spoj do jedne tabulky naprklad pomoc databazoveho dotazu, cmz je cesta k aplikaci siroke nabdky jednorelacnch DZD nastroju otevrena. Protoze velikost vysledne tabulky roste obecne velmi rychle s poctem spojovanych relac, jsou do uziteho databazoveho dotazu obycejne zacleneny omezujc podmnky vychazejc z intuice uzivatele. V tomto prstupu vsak nelze vetsinou rozmery vysledne relace predem odhadnout a ani snadno urcit, zda se uzitym typem spojen neztrac z dat nektera dulezita cast. Uziva

relational knowledge bases

Abstract: Predictive modelling and classification problems are central analytical tasks in Customer Relationship Management (CRM). Analysts typically do not have information on a customer’s social network available. Phone carriers are an exception, where companies accumulate huge amounts of telephone calling records providing information not only about the usage behaviour of a single customer, but also about how customers interact with each other. In this paper, we want to improve predictive modeling tasks leveraging social network information and respective network effects. The analysis of huge amounts of call detail data, which exhibit a graph structure poses new challenges for predictive modelling. We focus on Markov logic networks as one of the most promising recent developments in the field of multi-relational data mining. We will provide an experimental evaluation of Markov logic networks, propositional learners and a propositionalization approach. The experiments are based on a dataset from a European telecommunication provider.

• No classi cation problem No class given or No interest in class • Instead trying to nd description of (groups) of instances or relationship between di erent attributes and their values

One xed target vs Disjunction of all attributes Rule learning vs Arbitrary instance assignment Overlap vs Non-overlap • Similarities: Finding groups of instances showing unexpected behavior with regard to target attribute(s) Heuristic approach since number of possible partitions too large to exhaust 2. Getting away from the heuristic