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

Unification in Description Logics has been proposed as an inference service that can, for example, be used to detect redundancies in ontologies. For the Description Logic $\mathcal{EL}$, which is used to define several large biomedical ontologies, unification is NP-complete. An NP unification algorithm for $\mathcal{EL}$ based on a translation into propositional satisfiability (SAT) has recently been presented. In this paper, we extend this SAT encoding in two directions: on the one hand, we add general concept inclusion axioms, and on the other hand, we add role hierarchies ($\mathcal{H}$) and transitive roles (R+). For the translation to be complete, however, the ontology needs to satisfy a certain cycle restriction. The SAT translation depends on a new rewriting-based characterization of subsumption w.r.t. $\mathcal{ELH}_{{R}^+}$-ontologies.

SAT encoding of unification in ELH R+ w.r.t. cycle-restricted ontologies

Concrete domains have been introduced in Description Logics (DLs) to enable reference to concrete objects (such as numbers) and predefined predicates on these objects (such as numerical comparisons) when defining concepts. To retain decidability when integrating a concrete domain into a decidable DL, the domain must satisfy quite strong restrictions. In previous work, we have analyzed the most prominent such condition, called $\omega $-admissibility, from an algebraic point of view. This provided us with useful algebraic tools for proving $\omega $-admissibility, which allowed us to find new examples for concrete domains whose integration leaves the prototypical expressive DL $\mathcal {ALC}$ decidable.

An Algebraic View on p-Admissible Concrete Domains for Lightweight Description Logics

Intelligent access to heterogeneous information sources: report on the 4th workshop on knowledge representation meets databases

Extending the Description Logic with Acyclic TBoxes.

Storage and transfer of information as well as interfaces for accessing this information have undergone a remarkable evolution. Nevertheless, information systems are still not "intelligent" in the sense that they "understand" the information they store, manipulate, and present to their users. A case in point is the world wide web and search engines allowing to access the vast amount of information available there. Web-pages are mostly written for human consumption and the markup provides only rendering information for textual and graphical information. Search engines are usually based on keyword search and often provide a huge number of answers, many of which are completely irrelevant, whereas some of the more interesting answers are not found. In contrast, the vision of a "semantic web" aims for machine-understandable web resources, whose content can then be comprehended and processed both by automated tools, such as search engines, and by human users. The content-based representation of information requires representation formalisms with a well-defined formal semantics since otherwise there cannot be a common understanding of the represented information. This semantics can elegantly be provided by a translation into an appropriate logic or the use of a logic-based formalism in the first place. This logical approach has the additional advantage that logical inferences can then be used to reason about the represented information, thus detecting inconsistencies and computing implicit information. However, in this setting there is a fundamental tradeoff between the expressivity of the representation formalism on the one hand, and the efficiency of reasoning with this formalism on the other hand.This motivates the "engineering of logics", i.e., the design of logical formalisms that are tailored to specific representation tasks. This also encompasses the formal investigation of the relevant inference problems, the development of appropriate inferences procedures, and their implementation, optimization, and empirical evaluation. Another important topic in this context is the combination of logics and their inference procedures since a given application my require the use of more than one specialized logic. The talk will illustrate this approach with the example of so-called Description Logics and their application as ontology languages for the semantic web.

/pdf/engineering-of-logics-for-the-content-based-representation-3p0z14an5r.pdf

Franz Baader

Papers

SAT encoding of unification in ELH R+ w.r.t. cycle-restricted ontologies

An Algebraic View on p-Admissible Concrete Domains for Lightweight Description Logics

Intelligent access to heterogeneous information sources: report on the 4th workshop on knowledge representation meets databases

Extending the Description Logic with Acyclic TBoxes.

Engineering of Logics for the Content-Based Representation of Information