Ever since the early days of Artificial Intelligence and the development of the first knowledge-based systems in the 70s people have dreamt of self-learning machines. When knowledge-based systems grew larger and the commercial interest in these technologies increased, people became aware of the knowledge acquisition bottleneck and the necessity to (partly) automatize the creation and maintenance of knowledge bases. Today, many applications which exhibit ’intelligent’ behavior thanks to symbolic knowledge representation and logical inference rely on ontologies and the standards provided by the World Wide Web Committee (W3C). Supporting the construction of ontologies and populating them with instantiations of both concepts and relations, commonly referred to as ontology learning. Early research in ontology learning has concentrated on the extraction of facts or schema-level knowledge from textual resources building upon earlier work in the field of computational linguistics and lexical acquisition. However, as we will show in this book, ontology learning is a very diverse and interdisciplinary field of research. Ontology learning approaches are as heterogeneous as the sources of data on the web, and as different from one another as the types of knowledge representations called “ontologies”. In the remainder of this introduction, we briefly summarize the state-of-the-art in ontology learning and elaborate on what we consider as the key challenges for current and future ontology learning research. An Introduction to Ontology Learning

Advertisements