Artificial intelligence and machine learning are fields of research that have become very popular and are getting more attention in the media as our computational power increases and the theories and latest developments of these fields can be put into practice in the real world. The field of machine learning consists of different paradigms, two of which are the symbolic and connectionist paradigms. In 1991 it was pointed out by Minsky that we could benefit from sharing ideas between the paradigms instead of competing for dominance in the field. That is why this thesis is investigating two approaches to inductive logic programming, where the main research goals are to, first: find similarities or differences between the approaches and potential areas where cross-pollination could be beneficial, and secondly: investigate their relative performance to each other based on the results published in the research. The approaches investigated are Meta-Interpretive Learning and Inductive Metalogic Programming, which belong to the symbolic paradigm of machine learning. The research is conducted through a comparative study based on published research papers. The conclusion to the study suggests that at least two aspects of the approaches could potentially be shared between them, namely the reversible aspect of the meta-interpreter and restricting the hypothesis space using the Herbrand base. However, the findings regarding performance were deemed incompatible, in terms of a fair one to one comparison. The results of the study are mainly specific, but could be interpreted as motivation for similar collaboration efforts between different paradigms.

Multi Relational Data Mining has become popular due to the limitations of propositional problem definition in structured domains and the tendency of storing data in relational databases. However, as patterns involve multiple relations, the search space of possible hypothesis becomes intractably complex. In order to cope with this problem, several relational knowledge discovery systems have been developed employing various search strategies, heuristics and language pattern limitations. In this thesis, Inductive Logic Programming (ILP) based concept discovery is studied and two systems based on a hybrid methodology employing ILP and APRIORI, namely Confidence-based Concept Discovery and Concept Rule Induction System, are proposed. In Confidence-based Concept Discovery and Concept Rule Induction System, the main aim is to relax the strong declarative biases and user-defined specifications. Moreover, this new method directly works on relational databases. In addition to this, the traditional definition of confidence from relational database perspective is modified to express Closed World Assumption in first-order logic. A new confidence-based pruning method based on the improved definition is applied in the APRIORI lattice. Moreover, a new hypothesis evaluation criterion is used for expressing the quality of patterns in the search space. In addition to this, in Concept Rule Induction System, the constructed rule quality is further improved by using an improved generalization metod. Finally, a set of experiments are conducted on real-world problems to evaluate the performance of the proposed method with similar systems in terms of support and confidence.

Ο Επαγωγικός Λογικός Προγραμματισμός (Inductive Logic Programming ή, σε συντομογραφία ILP) είναι ο ερευνητικός τομέας της Τεχνητής Νοημοσύνης (Artificial Intelligence) που δραστηριοποιείται στη τομή των γνωστικών περιοχών της Μάθησης Μηχανής (Machine Learning) και του Λογικού Προγραμματισμού (Logic Programming).Ο όρος επαγωγικός εκφράζει την ιδέα του συλλογισμού από το επί μέρους στο γενικό. Μέσω της επαγωγικής μάθησης μηχανής ο ILP επιτυγχάνει το στόχο του που είναι η δημιουργία εργαλείων και η ανάπτυξη τεχνικών για την εξαγωγή υποθέσεων από παρατηρήσεις (παραδείγματα) και η σύνθεση-απόκτηση νέας γνώσης από εμπειρικές παρατηρήσεις. Σε αντίθεση με της περισσότερες άλλες προσεγγίσεις της επαγωγικής μάθησης ο ILP ενδιαφέρεται για της ιδιότητες του συμπερασμού με κανόνες για την σύγκλιση αλγορίθμων και για την υπολογιστική πολυπλοκότητα των διαδικασιών. Ο ILP ασχολείται με την ανάπτυξη τεχνικών και εργαλείων για την σχεσιακή ανάλυση δεδομένων. Εφαρμόζεται απευθείας σε δεδομένα πολλαπλών συσχετισμών για την ανακάλυψη προτύπων. Τα πρότυπα που ανακαλύπτονται από τα συστήματα στον ILP εκφράζονται ως λογικά προγράμματα. Τα λογικά προγράμματα αποτελούνται από ειδικούς κανόνες, οι οποίοι χωρίζονται στις προϋποθέσεις και στα συμπεράσματα. Ο ILP έχει χρησιμοποιηθεί εκτεταμένα σε προβλήματα που αφορούν τη μοριακή βιολογία, την βιοχημεία και την χημεία. Τα παραδείγματα, οι κανόνες εκφράζουν την γνώση υποβάθρου εκφράζονται σε μια γλώσσα λογικού προγραμματισμού όπως η Prolog. Ο Επαγωγικός Λογικός Προγραμματισμός διαφοροποιείται από τις άλλες μορφές Μάθησης Μηχανής, αφ’ ενός μεν λόγω της χρήσης μιας εκφραστικής γλώσσας αναπαράστασης και αφ’ ετέρου από τη δυνατότητά του να χρησιμοποιεί τη γνώση υποβάθρου. Διάφορες εφαρμογές έχουν αναπτυχθεί, εκ των οποίων η πιο πρόσφατη είναι η Progol, που αποτελείται από ένα διερμηνέα της Prolog ο οποίος συνοδεύεται από έναν αλγόριθμο Αντίστροφης Συνεπαγωγής (Inverse Entailment) που κατασκευάζει νέες προτάσεις με τη γενίκευση των παραδειγμάτων που περιέχονται στη βάση δεδομένων της Prolog. Η θεωρία του Επαγωγικού Λογικού Προγραμματισμού εγγυάται ότι η Progol θα διεξάγει μια αποδεκτή αναζήτηση στο διάστημα των γενικεύσεων, βρίσκοντας το ελάχιστο σύνολο προτάσεων, από το οποίο όλα τα παραδείγματα μπορούν να προκύψουν. Στην συγκεκριμένη εργασία η Progol είναι το εργαλείο που χρησιμοποιείται για την ανάπτυξη των παραδειγμάτων εφαρμογής του ILP. / The Inductive Reasonable Planning (Inductive Logic Programming or, in abbreviation ILP) is the inquiring sector Artificial Intelligence that is activated in the section of cognitive regions of Learning of Machine (Machine Learning) and Reasonable Planning (Logic Programming). The term inductive expresses the idea of reasoning from on part in general. Via the inductive learning of machine the ILP achieves his objective that is the creation of tools and the growth of techniques for the export of affairs from observations (examples) and composition of new knowledge from empiric observations. Contrary to more other approaches of inductive learning the ILP is interested for its inference attributes with rules on the convergence of algorithms and on the calculating complexity of processes. The ILP deals with the growth of techniques and tools for the relational analysis of data. It is applied directly in data of multiple correlations on the discovery of models. The models that are discovered by the systems in the ILP are expressed as reasonable programs. The reasonable programs are constituted by special rules, which are separated in the conditions and in the conclusions. The ILP has been used extensive in problems that concern the molecular biology, the biochemistry and the chemistry. The examples, the rules express the knowledge of background are expressed in a language of reasonable planning as the Prolog. The Inductive Reasonable Planning is differentiated by the other forms of Learning of Machine, on the one hand men because the use of expressive language of representation and on the other hand by his possibility of using the knowledge of background. Various applications have been developed, from which most recent is Progol, that is constituted from interpreter of Prolog which is accompanied by a algorithm of Inverse Entailment that manufactures new proposals with the generalisation of examples that is contained in the base of data of Prolog. theory of Inductive Reasonable Planning guarantees that the Progol will carry out a acceptable search in the interval of generalisations, finding the minimal total of proposals, from which all the examples can result. In the particular work the Progol is the tool that is used for the growth of examples of application of ILP.

abstract: Artificial general intelligence consists of many components, one of which is Natural Language Understanding (NLU). One of the applications of NLU is Reading Comprehension where it is expected that a system understand all aspects of a text. Further, understanding natural procedure-describing text that deals with existence of entities and effects of actions on these entities while doing reasoning and inference at the same time is a particularly difficult task. A recent natural language dataset by the Allen Institute of Artificial Intelligence, ProPara, attempted to address the challenges to determine entity existence and entity tracking in natural text. As part of this work, an attempt is made to address the ProPara challenge. The Knowledge Representation and Reasoning (KRR) community has developed effective techniques for modeling and reasoning about actions and similar techniques are used in this work. A system consisting of Inductive Logic Programming (ILP) and Answer Set Programming (ASP) is used to address the challenge and achieves close to state-of-the-art results and provides an explainable model. An existing semantic role label parser is modified and used to parse the dataset. On analysis of the learnt model, it was found that some of the rules were not generic enough. To overcome the issue, the Proposition Bank dataset is then used to add knowledge in an attempt to generalize the ILP learnt rules to possibly improve the results. / Dissertation/Thesis / Masters Thesis Computer Science 2019

A combinação de lógica e probabilidade (lógicas probabilísticas) tem sido um tópico bastante estudado nas últimas décadas. A maioria de propostas para estes formalismos pressupõem que tanto as sentenças lógicas como as probabilidades sejam especificadas por especialistas. Entretanto, a crescente disponibilidade de dados relacionais sugere o uso de técnicas de aprendizado de máquina para produzir sentenças lógicas e estimar probabilidades. Este trabalho apresenta contribuições em termos de representação de conhecimento e aprendizado. Primeiro, uma linguagem lógica probabilística de primeira ordem é proposta. Em seguida, três algoritmos de aprendizado de lógica de descrição probabilística crALC são apresentados: um algoritmo probabilístico com ênfase na indução de sentenças baseada em classificadores Noisy-OR; um algoritmo que foca na indução de inclusões probabilísticas (componente probabilístico de crALC); um algoritmo de natureza probabilística que induz sentenças lógicas ou inclusões probabilísticas. As propostas de aprendizado são avaliadas em termos de acurácia em duas tarefas: no aprendizado de lógicas de descrição e no aprendizado de terminologias probabilísticas em crALC. Adicionalmente, são discutidas aplicações destes algoritmos em processos de recuperação de informação: duas abordagens para extensão semântica de consultas na Web usando ontologias probabilísticas são discutidas. / The combination of logic and probabilities (probabilistic logics) is a topic that has been extensively explored in past decades. The majority of work in probabilistic logics assumes that both logical sentences and probabilities are specified by experts. As relational data is increasingly available, machine learning algorithms have been used to induce both logical sentences and probabilities. This work contributes in knowledge representation and learning. First, a rst-order probabilistic logic is proposed. Then, three algorithms for learning probabilistic description logic crALC are given: a probabilistic algorithm focused on learning logical sentences and based on Noisy-OR classiers; an algorithm that aims at learning probabilistic inclusions (probabilistic component of crALC) and; an algorithm that using a probabilistic setting, induces either logical sentences or probabilistic inclusions. Evaluation of these proposals has been performed in two situations: by measuring learning accuracy of both description logics and probabilistic terminologies. In addition, these learning algorithms have been applied to information retrieval processes: two approaches for semantic query extension through probabilistic ontologies are discussed.

Le Web sémantique propose un ensemble de standards et d'outils pour la formalisation et l'interopérabilité de connaissances partagées sur le Web, sous la forme d'ontologies. Les ontologies biomédicales et les données associées constituent de nos jours un ensemble de connaissances complexes, hétérogènes et interconnectées, dont l'analyse est porteuse de grands enjeux en santé, par exemple dans le cadre de la pharmacovigilance. On proposera dans cette thèse des méthodes permettant d'utiliser ces ontologies biomédicales pour étendre les possibilités d'un processus de fouille de données, en particulier, permettant de faire cohabiter et d'exploiter les connaissances de plusieurs ontologies biomédicales. Les travaux de cette thèse concernent dans un premier temps une méthode fondée sur les structures de patrons, une extension de l'analyse formelle de concepts pour la découverte de co-occurences de événements indésirables médicamenteux dans des données patients. Cette méthode utilise une ontologie de phénotypes et une ontologie de médicaments pour permettre la comparaison de ces événements complexes, et la découverte d'associations à différents niveaux de généralisation, par exemple, au niveau de médicaments ou de classes de médicaments. Dans un second temps, on utilisera une méthode numérique fondée sur des mesures de similarité sémantique pour la classification de déficiences intellectuelles génétiques. On étudiera deux mesures de similarité utilisant des méthodes de calcul différentes, que l'on utilisera avec différentes combinaisons d'ontologies phénotypiques et géniques. En particulier, on quantifiera l'influence que les différentes connaissances de domaine ont sur la capacité de classification de ces mesures, et comment ces connaissances peuvent coopérer au sein de telles méthodes numériques. Une troisième étude utilise les données ouvertes liées ou LOD du Web sémantique et les ontologies associées dans le but de caractériser des gènes responsables de déficiences intellectuelles. On utilise ici la programmation logique inductive, qui s'avère adaptée pour fouiller des données relationnelles comme les LOD, en prenant en compte leurs relations avec les ontologies, et en extraire un modèle prédictif et descriptif des gènes responsables de déficiences intellectuelles. L'ensemble des contributions de cette thèse montre qu'il est possible de faire coopérer avantageusement une ou plusieurs ontologies dans divers processus de fouille de données / The semantic Web proposes standards and tools to formalize and share knowledge on the Web, in the form of ontologies. Biomedical ontologies and associated data represents a vast collection of complex, heterogeneous and linked knowledge. The analysis of such knowledge presents great opportunities in healthcare, for instance in pharmacovigilance. This thesis explores several ways to make use of this biomedical knowledge in the data mining step of a knowledge discovery process. In particular, we propose three methods in which several ontologies cooperate to improve data mining results. A first contribution of this thesis describes a method based on pattern structures, an extension of formal concept analysis, to extract associations between adverse drug events from patient data. In this context, a phenotype ontology and a drug ontology cooperate to allow a semantic comparison of these complex adverse events, and leading to the discovery of associations between such events at varying degrees of generalization, for instance, at the drug or drug class level. A second contribution uses a numeric method based on semantic similarity measures to classify different types of genetic intellectual disabilities, characterized by both their phenotypes and the functions of their linked genes. We study two different similarity measures, applied with different combinations of phenotypic and gene function ontologies. In particular, we investigate the influence of each domain of knowledge represented in each ontology on the classification process, and how they can cooperate to improve that process. Finally, a third contribution uses the data component of the semantic Web, the Linked Open Data (LOD), together with linked ontologies, to characterize genes responsible for intellectual deficiencies. We use Inductive Logic Programming, a suitable method to mine relational data such as LOD while exploiting domain knowledge from ontologies by using reasoning mechanisms. Here, ILP allows to extract from LOD and ontologies a descriptive and predictive model of genes responsible for intellectual disabilities. These contributions illustrates the possibility of having several ontologies cooperate to improve various data mining processes

A comunicação multilíngue é uma tarefa cada vez mais imperativa no cenário atual de grande disseminação de informações em diversas línguas. Nesse contexto, são de grande relevância os sistemas de tradução automática, que auxiliam tal comunicação, automatizando-a. Apesar de ser uma área de pesquisa bastante antiga, a Tradução Automática ainda apresenta muitos problemas. Um dos principais problemas é a ambigüidade lexical, ou seja, a necessidade de escolha de uma palavra, na língua alvo, para traduzir uma palavra da língua fonte quando há várias opções de tradução. Esse problema se mostra ainda mais complexo quando são identificadas apenas variações de sentido nas opções de tradução. Ele é denominado, nesse caso, \"ambigüidade lexical de sentido\". Várias abordagens têm sido propostas para a desambiguação lexical de sentido, mas elas são, em geral, monolíngues (para o inglês) e independentes de aplicação. Além disso, apresentam limitações no que diz respeito às fontes de conhecimento que podem ser exploradas. Em se tratando da língua portuguesa, em especial, não há pesquisas significativas voltadas para a resolução desse problema. O objetivo deste trabalho é a proposta e desenvolvimento de uma nova abordagem de desambiguação lexical de sentido, voltada especificamente para a tradução automática, que segue uma metodologia híbrida (baseada em conhecimento e em córpus) e utiliza um formalismo relacional para a representação de vários tipos de conhecimentos e de exemplos de desambiguação, por meio da técnica de Programação Lógica Indutiva. Experimentos diversos mostraram que a abordagem proposta supera abordagens alternativas para a desambiguação multilíngue e apresenta desempenho superior ou comparável ao do estado da arte em desambiguação monolíngue. Adicionalmente, tal abordagem se mostrou efetiva como mecanismo auxiliar para a escolha lexical na tradução automática estatística / Crosslingual communication has become a very imperative task in the current scenario with the increasing amount of information dissemination in several languages. In this context, machine translation systems, which can facilitate such communication by providing automatic translations, are of great importance. Although research in Machine Translation dates back to the 1950\'s, the area still has many problems. One of the main problems is that of lexical ambiguity, that is, the need for lexical choice when translating a source language word that has several translation options in the target language. This problem is even more complex when only sense variations are found in the translation options, a problem named \"sense ambiguity\". Several approaches have been proposed for word sense disambiguation, but they are in general monolingual (for English) and application-independent. Moreover, they have limitations regarding the types of knowledge sources that can be exploited. Particularly, there is no significant research aiming to word sense disambiguation involving Portuguese. The goal of this PhD work is the proposal and development of a novel approach for word sense disambiguation which is specifically designed for machine translation, follows a hybrid methodology (knowledge and corpus-based), and employs a relational formalism to represent various kinds of knowledge sources and disambiguation examples, by using Inductive Logic Programming. Several experiments have shown that the proposed approach overcomes alternative approaches in multilingual disambiguation and achieves higher or comparable results to the state of the art in monolingual disambiguation. Additionally, the approach has shown to effectively assist lexical choice in a statistical machine translation system

Application of neural networks in deep learning is rapidly growing due to their ability to outperform other machine learning algorithms in different kinds of problems. But one big disadvantage of deep neural networks is its internal logic to achieve the desired output or result that is un-understandable and unexplainable. This behavior of the deep neural network is known as “black box”. This leads to the following questions: how prevalent is the black box problem in the research literature during a specific period of time? The black box problems are usually addressed by socalled rule extraction. The second research question is: what rule extracting methods have been proposed to solve such kind of problems? To answer the research questions, a systematic literature review was conducted for data collection related to topics, the black box, and the rule extraction. The printed and online articles published in higher ranks journals and conference proceedings were selected to investigate and answer the research questions. The analysis unit was a set of journals and conference proceedings articles related to the topics, the black box, and the rule extraction. The results conclude that there has been gradually increasing interest in the black box problems with the passage of time mainly because of new technological development. The thesis also provides an overview of different methodological approaches used for rule extraction methods.

Ο Επαγωγικός Λογικός Προγραμματισμός (Inductive Logic Programming ή, σε συντομογραφία ILP) είναι ο ερευνητικός τομέας της Τεχνητής Νοημοσύνης (Artificial Intelligence) που δραστηριοποιείται στη τομή των γνωστικών περιοχών της Μάθησης Μηχανής (Machine Learning) και του Λογικού Προγραμματισμού (Logic Programming).Ο όρος επαγωγικός εκφράζει την ιδέα του συλλογισμού από το επί μέρους στο γενικό. Μέσω της επαγωγικής μάθησης μηχανής ο Επαγωγικός Λογικός Προγραμματισμός επιτυγχάνει το στόχο του που είναι η δημιουργία εργαλείων και η ανάπτυξη τεχνικών για την εξαγωγή υποθέσεων από παρατηρήσεις (παραδείγματα) και η σύνθεση-απόκτηση νέας γνώσης από εμπειρικές παρατηρήσεις. Σε αντίθεση με της περισσότερες άλλες προσεγγίσεις της επαγωγικής μάθησης ο Επαγωγικός Λογικός Προγραμματισμός ενδιαφέρεται για της ιδιότητες του συμπερασμού με κανόνες για την σύγκλιση αλγορίθμων και για την υπολογιστική πολυπλοκότητα των διαδικασιών. Ο Επαγωγικός Λογικός Προγραμματισμός ασχολείται με την ανάπτυξη τεχνικών και εργαλείων για την σχεσιακή ανάλυση δεδομένων. Εφαρμόζεται απευθείας σε δεδομένα πολλαπλών συσχετισμών για την ανακάλυψη προτύπων. Τα πρότυπα που ανακαλύπτονται από τα συστήματα στον Επαγωγικό Λογικό Προγραμματισμό προκύπτουν από κάποιο γνωστό θεωρητικό υπόβαθρο και θετικά και αρνητικά παραδείγματα και εκφράζονται ως λογικά προγράμματα. Ο Επαγωγικός Λογικός Προγραμματισμός έχει χρησιμοποιηθεί εκτεταμένα σε προβλήματα που αφορούν τη μοριακή βιολογία, την βιοχημεία και την χημεία. Ο Επαγωγικός Λογικός Προγραμματισμός διαφοροποιείται από τις άλλες μορφές Μάθησης Μηχανής, αφ’ ενός μεν λόγω της χρήσης μιας εκφραστικής γλώσσας αναπαράστασης και αφ’ ετέρου από τη δυνατότητά του να χρησιμοποιεί τη γνώση υποβάθρου. Έχουν αναπτυχθεί διάφορες μηχανισμούς υλοποίησης του ILP, εκ των οποίων η πιο πρόσφατη είναι η Progol, που βασίζεται σε ένα διερμηνέα της Prolog ο οποίος συνοδεύεται από έναν αλγόριθμο Αντίστροφης Συνεπαγωγής (Inverse Entailment). Η Progol κατασκευάζει νέες προτάσεις με τη γενίκευση των παραδειγμάτων που περιέχονται στη βάση δεδομένων που της δίνεται. Η θεωρία του Επαγωγικού Λογικού Προγραμματισμού εγγυάται ότι η Progol θα διεξάγει μια αποδεκτή αναζήτηση στο διάστημα των γενικεύσεων, βρίσκοντας το ελάχιστο σύνολο προτάσεων, από το οποίο όλα τα παραδείγματα μπορούν να προκύψουν. Σε αυτή την εργασία θα αναπτυχθούν αναλυτικά η θεωρία και οι κανόνες του Επαγωγικού Λογικού Προγραμματισμού, τα είδη των προβλημάτων που επιλύονται μέσω του Επαγωγικού Λογικού Προγραμματισμού, οι μέθοδοι που ακολουθούνται καθώς και ο τρόπος με τον οποίο αναπτύσσονται οι εφαρμογές του Επαγωγικού Λογικού Προγραμματισμού. Θα δοθούν επίσης παραδείγματα κατάλληλα για την κατανόηση των γνώσεων αυτών από ένα ακροατήριο που διαθέτει βασικές γνώσεις Λογικής και Λογικού Προγραμματισμού. / Inductive Logic Programming is a research area of Artificial Intelligence that operates in the intersection of cognitive areas of Machine Learning and Logic Programming. Through inductive machine learning, Inductive Logic Programming‟s objective is creating tools and developing techniques to extract new knowledge composing a background one and empirical observations (examples). Some methods are employed, the best known of which is the reverse implication, the reverse resolution and the inverse implication. Based on Inductive Logic Programming, some systems have been developed for knowledge production. The most widely used system is Progol, which uses an input of examples and background knowledge, whichε is stated in a kind of grammar compatible to that the programming language Prolog, and generates procedures in the same language that illustrate these examples. Other systems are FOIL, MOBAL, GOLEM and LINUS. There is also Cigol which is a programming language based on the theory of Inductive Logic Programming. These systems are used in many applications. The most important is the area of pharmacology, such as predictive toxicology, the provision of rheumatic disease and the design of drugs for Alzheimer's. Applications can also be found in programming, linguistics and games like chess.

A combinação de lógica e probabilidade (lógicas probabilísticas) tem sido um tópico bastante estudado nas últimas décadas. A maioria de propostas para estes formalismos pressupõem que tanto as sentenças lógicas como as probabilidades sejam especificadas por especialistas. Entretanto, a crescente disponibilidade de dados relacionais sugere o uso de técnicas de aprendizado de máquina para produzir sentenças lógicas e estimar probabilidades. Este trabalho apresenta contribuições em termos de representação de conhecimento e aprendizado. Primeiro, uma linguagem lógica probabilística de primeira ordem é proposta. Em seguida, três algoritmos de aprendizado de lógica de descrição probabilística crALC são apresentados: um algoritmo probabilístico com ênfase na indução de sentenças baseada em classificadores Noisy-OR; um algoritmo que foca na indução de inclusões probabilísticas (componente probabilístico de crALC); um algoritmo de natureza probabilística que induz sentenças lógicas ou inclusões probabilísticas. As propostas de aprendizado são avaliadas em termos de acurácia em duas tarefas: no aprendizado de lógicas de descrição e no aprendizado de terminologias probabilísticas em crALC. Adicionalmente, são discutidas aplicações destes algoritmos em processos de recuperação de informação: duas abordagens para extensão semântica de consultas na Web usando ontologias probabilísticas são discutidas. / The combination of logic and probabilities (probabilistic logics) is a topic that has been extensively explored in past decades. The majority of work in probabilistic logics assumes that both logical sentences and probabilities are specified by experts. As relational data is increasingly available, machine learning algorithms have been used to induce both logical sentences and probabilities. This work contributes in knowledge representation and learning. First, a rst-order probabilistic logic is proposed. Then, three algorithms for learning probabilistic description logic crALC are given: a probabilistic algorithm focused on learning logical sentences and based on Noisy-OR classiers; an algorithm that aims at learning probabilistic inclusions (probabilistic component of crALC) and; an algorithm that using a probabilistic setting, induces either logical sentences or probabilistic inclusions. Evaluation of these proposals has been performed in two situations: by measuring learning accuracy of both description logics and probabilistic terminologies. In addition, these learning algorithms have been applied to information retrieval processes: two approaches for semantic query extension through probabilistic ontologies are discussed.