ONTOLOGY LEARNING and the HUMANITIES

This chapter reviews the current state of play in the use of ontologies in the humanities, with best-practice examples from selected disciplines. It looks at the specific domain problems faced by the humanities, and examines the various approaches currently being employed to construct, maintain, and develop humanities ontologies. The application of ontology learning in the humanities is discussed by reviewing a range of research projects in different disciplines. The chapter concludes with an assessment of the future potential of ontology learning in the humanities, and an attempt to set out a research agenda for this field.

Learning SKOS Relations for Terminological Ontologies from Text

The problem of learning concept hierarchies and terminological ontologies can be divided into two sub-tasks: concept extraction and relation learning. The authors of this chapter describe a novel approach to learn relations automatically from unstructured text corpus based on probabilistic topic models. The authors provide definition (Information Theory Principle for Concept Relationship) and quantitative measure for establishing “broader” (or “narrower”) and “related” relations between concepts. They present a relation learning algorithm to automatically interconnect concepts into concept hierarchies and terminological ontologies with the probabilistic topic models learned. In this experiment, around 7,000 ontology statements expressed in terms of “broader” and “related” relations are generated using different combination of model parameters. The ontology statements are evaluated by domain experts and the results show that the highest precision of the learned ontologies is around 86.6% and structures of learned ontologies remain stable when values of the parameters are changed in the ontology learning algorithm.

Cross-language Ontology Learning

A formal ontology does not contain lexical knowledge; it is by nature language-independent. Mappings can be added between the ontology and, arbitrarily, many lexica in any number of languages. The result of this operation is what is here referred to as a cross-language ontology. A cross-language ontology can be a useful resource for machine translation or cross-language information retrieval. This chapter focuses on ways of automatically building an ontology by exploiting cross-language information from parallel corpora. The goal is to improve the automatic learning results compared to learning an ontology from resources in a single language. The authors present a framework for cross-language ontology learning, providing a setting in which cross-language evidence (data) can be integrated and quantified. The aim is to investigate the following question: Can cross-language data teach us more than data from a single language for the ontology learning task?

Ontology-Based Knowledge Capture and Sharing in Enterprise Organisations

Despite years of effort in building organisational taxonomies, the potential of ontologies to support knowledge management in complex technical domains is under-exploited. The authors of this chapter present an approach to using rich domain ontologies to support sense-making tasks associated with resolving mechanical issues. Using Semantic Web technologies, the authors have built a framework and a suite of tools which support the whole semantic knowledge lifecycle. These are presented by describing the process of issue resolution for a simulated investigation concerning failure of bicycle brakes. Foci of the work have included ensuring that semantic tasks fit in with users’ everyday tasks, to achieve user acceptability and support the flexibility required by communities of practice with differing local sub-domains, tasks, and terminology.

Named Entity Recognition for Ontology Population using Background Knowledge from Wikipedia

Named Entity Recognition (NER) deals with identifying and classifying atomic texts into pre-defined ontological classes. It is the enabling technique to many complex knowledge acquisition tasks. The recent flourish of Web resources has opened new opportunities and challenges for knowledge acquisition. In the domain of NER and its application in ontology population, considerable research work has been dedicated to exploiting background knowledge from Web resources to enhance the accuracy of the system. This chapter gives a review of existing literature in this domain with an emphasis on using background knowledge extracted from the Web resources. The authors discuss the benefits of using background knowledge and the inadequacies of existing work. They then propose a novel method that automatically creates domain-specific background knowledge by exploring the Wikipedia knowledge base in a domain- and language-independent way. The authors empirically show that the method can be adapted to ontology population, and generates high quality background knowledge that improves the accuracy of domain-specific NER.

A Cognitive-Based Approach to Identify Topics in Text Using the Web as a Knowledge Source

This chapter explores the problem of topic identification from text. It is first argued that the conventional representation of text as bag-of-words vectors will always have limited success in arriving at the underlying meaning of text until the more fundamental issues of feature independence in vector-space and ambiguity of natural language are addressed. Next, a groundbreaking approach to text representation and topic identification that deviates radically from current techniques used for document classification, text clustering, and concept discovery is proposed. This approach is inspired by human cognition, which allows ‘meaning’ to emerge naturally from the activation and decay of unstructured text information retrieved from the Web. This paradigm shift allows for the exploitation rather than avoidance of dependence between terms to derive meaning without the complexity introduced by conventional natural language processing techniques. Using the unstructured texts in Web pages as a source of knowledge alleviates the laborious handcrafting of formal knowledge bases and ontologies that are required by many existing techniques. Some initial experiments have been conducted, and the results are presented in this chapter to illustrate the power of this new approach.

Evidence Sources, Methods and Use Cases for Learning Lightweight Domain Ontologies

By providing interoperability and shared meaning across actors and domains, lightweight domain ontologies are a cornerstone technology of the Semantic Web. This chapter investigates evidence sources for ontology learning and describes a generic and extensible approach to ontology learning that combines such evidence sources to extract domain concepts, identify relations between the ontology’s concepts, and detect relation labels automatically. An implementation illustrates the presented ontology learning and relation labeling framework and serves as the basis for discussing possible pitfalls in ontology learning. Afterwards, three use cases demonstrate the usefulness of the presented framework and its application to real-world problems.

GO-Based Term Semantic Similarity

As the Gene Ontology (GO) plays more and more important roles in bioinformatics research, there has been great interest in developing objective and accurate methods for calculating semantic similarity between GO terms. In this chapter, the authors first introduce the basic concepts related to the GO and then briefly review the current advances and challenges in the development of methods for calculating semantic similarity between GO terms. Then, the authors introduce a semantic similarity method that does not rely on external data sources. Using this method as an example, the authors show how different properties of the GO can be explored to calculate semantic similarities between pairs of GO terms. The authors conclude the chapter by presenting some thoughts on the directions for future research in this field.

Incorporating Correlations among Gene Ontology Terms into Predicting Protein Functions

The authors describe a new strategy that has better prediction performance than previous methods, which gives additional insights about the importance of the dependence between functional terms when inferring protein function.

Topic Extraction for Ontology Learning

This chapter addresses the issue of topic extraction from text corpora for ontology learning. The first part provides an overview of some of the most significant solutions present today in the literature. These solutions deal mainly with the inferior layers of the Ontology Learning Layer Cake. They are related to the challenges of the Terms and Synonyms layers. The second part shows how these pieces can be bound together into an integrated system for extracting meaningful topics. While the extracted topics are not proper concepts as yet, they constitute a convincing approach towards concept building and therefore ontology learning. This chapter concludes by discussing the research undertaken for filling the gap between topics and concepts as well as perspectives that emerge today in the area of topic extraction.