Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

We present YAGO, a light-weight and extensible ontology with high coverage and quality. YAGO builds on entities and relations and currently contains more than 1 million entities and 5 million facts. This includes the Is-A hierarchy as well as non-taxonomic relations between entities (such as HASONEPRIZE). The facts have been automatically extracted from Wikipedia and unified with WordNet, using a carefully designed combination of rule-based and heuristic methods described in this paper. The resulting knowledge base is a major step beyond WordNet: in quality by adding knowledge about individuals like persons, organizations, products, etc. with their semantic relationships - and in quantity by increasing the number of facts by more than an order of magnitude. Our empirical evaluation of fact correctness shows an accuracy of about 95%. YAGO is based on a logically clean model, which is decidable, extensible, and compatible with RDFS. Finally, we show how YAGO can be further extended by state-of-the-art information extraction techniques.

/pdf/yago-a-core-of-semantic-knowledge-13k3nmj6l2.pdf

Yago: a core of semantic knowledge

https://www.cs.ox.ac.uk/people/bernardo.cuencagrau/publications/PelletDemo.pdf

Pellet: A practical OWL-DL reasoner

Data integration is the problem of combining data residing at different sources, and providing the user with a unified view of these data. The problem of designing data integration systems is important in current real world applications, and is characterized by a number of issues that are interesting from a theoretical point of view. This document presents on overview of the material to be presented in a tutorial on data integration. The tutorial is focused on some of the theoretical issues that are relevant for data integration. Special attention will be devoted to the following aspects: modeling a data integration application, processing queries in data integration, dealing with inconsistent data sources, and reasoning on queries.

/pdf/data-integration-a-theoretical-perspective-58aesjvhuk.pdf

Data integration: a theoretical perspective

Formal Concept Analysis

The notions 'expressive power' or 'expressiveness' of knowledge representation languages (KR languages) can be found in most papers on knowledge representation; but these terms are usually just employed in an intuitive sense. The papers contain only informal descriptions of what is meant by expressiveness. There are several reasons that speak in favour of a formal definition of expressiveness: for example, if we want to show that certain expressions in one language cannot be expressed in another language, we need a strict formalism that can be used in mathematical proofs. Even though we shall only consider terminological KR languages—i.e. KR languages descending from the original system KL-ONE—in our motivation and in the examples, the definition of expressive power that will be given in this paper can be used for all KR languages with Tarski-style model-theoretic semantics. This definition will shed a new light on the tradeoff between expressiveness of a representation language and its computational tractability. There are KR languages with identical expressive power, but different complexity results for reasoning, which comes from the fact that sometimes the tradeoff lies between convenience and computational tractability. The definition of expressive power will be applied to compare various terminological KR languages known from the literature with respect to their expressiveness. This will yield examples for how to utilize the definition both in positive proofs— that is, proofs where it is shown that one language can be expressed by another language—and, more interestingly, in negative proofs—which show that a given language cannot be expressed by the other language.

A Formal Definition for the Expressive Power of Terminological Knowledge Representation Languages

Unification was originally introduced in automated deduction and term rewriting, but has recently also found applications in other fields. In this article, we give a survey of the results on unification obtained in two closely related, yet different, application areas of unification: description logics and modal logics.

/pdf/unification-in-modal-and-description-logics-10qx7qi401.pdf

Unification in modal and description logics

Motivated by the need for semantically well-founded and algorithmically managable formalisms for describing the functionality of Web services, we introduce an action formalism that is based on description logics (DLs), but is also rmly grounded on research in the reasoning about action community. Our main contribution is an analysis of how the choice of the DL inuences the complexity of standard reasoning tasks such as projection and executability, which are important for Web service discovery and composition.

/pdf/a-description-logic-based-approach-to-reasoning-about-web-3sw0r3vi0d.pdf

A Description Logic Based Approach to Reasoning about Web Services

We shall consider an application in mechanical engineering, and shall show that the adequate modelling of the terminology of this problem domain in a conventional concept language poses two main representation problems. The first requires access to concrete domains, such as real numbers, while the second asks for a construct which can be used to represent sequences of varying length. As shown in recent papers by the authors there exist extended concept languages—equipped with sound and complete reasoning algorithms—that satisfy the respective representation demands separately.

/pdf/extensions-of-concept-languages-for-a-mechanical-engineering-3xq2qkrkri.pdf

Extensions of Concept Languages for a Mechanical Engineering Application

Ontology-based data access (OBDA) generalizes query answering in databases towards deduction since (i) the fact base is not assumed to contain complete knowledge (i.e., there is no closed world assumption), and (ii) the interpretation of the predicates occurring in the queries is constrained by axioms of an ontology. OBDA has been investigated in detail for the case where the ontology is expressed by an appropriate Description Logic (DL) and the queries are conjunctive queries. Motivated by situation awareness applications, we investigate an extension of OBDA to the temporal case. As query language we consider an extension of the well-known propositional temporal logic LTL where conjunctive queries can occur in place of propositional variables, and as ontology language we use the prototypical expressive DL $\mathcal{ALC}$. For the resulting instance of temporalized OBDA, we investigate both data complexity and combined complexity of the query entailment problem.

/pdf/temporalizing-ontology-based-data-access-4z6yapgy3z.pdf

Franz Baader

Papers

A Formal Definition for the Expressive Power of Terminological Knowledge Representation Languages

Unification in modal and description logics

A Description Logic Based Approach to Reasoning about Web Services

Extensions of Concept Languages for a Mechanical Engineering Application

Temporalizing ontology-based data access