Creating ontological metadata for digital library content and services

Abstract

We use formal ontologies to represent knowledge about digital library content and services. Formal ontologies define concepts with logic in a frameinheritance structure. The expressiveness and precision of these structures supports computational reasoning that can be used in important ways. This paper focuses on the creation of ontological metadata.We create ontological content metadata by generating it from MARC (MAchine Readable Cataloging) data. MARC contains much information that is hard to exploit computationally. In particular, relationships between works are implicit in shared values and natural language notes. The conversion process involves specifying an ontological model, mapping MARC to the ontology, and reasoning about the data to create explicit links between works.Service metadata will be supplied by providers who wish to participate fully in a digital library that is implemented as a decentralized multi-agent system. Agents advertise by describing their services in terms of ontologically defined concepts. We reason about these descriptions to organize them into subsumption taxonomies. Agents can then find the best available services to meet their needs by describing their needs, without requiring a priori knowledge of other agents. This infrastructure has demonstrated its usefulness in a multi-agent system organized as a computational economy.

Agent Communication with Differentiated Ontologies: Eight New Measures of Description Compatibility

“…Broker agents know about services available in a community of agents, which are also represented by service descriptions. These descriptions may have been defined by agents as a means of advertising (Weinstein and Birmingham 1998). The broker applies some sort of filter to select a set of candidate services, then compares each candidate service description to the request.…”

“…We designed the representation of our artificial ontologies in accordance with our experience developing ontologies in complex domains (Weinstein and Birmingham 1998), and our expectations for the sensititivities of our measures of description compatibility. We knew that to match concept definitions we would convert concept definitions to description graphs (Borgida and Patel-Schneider 1994;Cohen and Hirsh 1994), so we thought in terms of graph structure.…”

“…These clustered correlations represent the structure of the world: for example, we expect tall men to also be relatively heavy. Or, say that we are concerned with agent services in digital libraries (Weinstein and Birmingham 1998), and OBJ00123 is one of a group of services that provide access to educational videos. We would expect that the intended audience of the videos and their conceptual-level would strongly covary, thus hypothetically supporting mappings between services for elementary-school, and those for reading-grade-level-2.…”

“…al. (Chawathe, Garcia-Molina et al 1994)), logic (e.g., Guha (Guha 1992)), or as sets of tables and procedures (e.g., Weinstein and Birmingham (Weinstein and Birmingham 1998)). Most often, mapping is between local terminologies and a global terminology, since this configuration requires a number of mappings linear in the number of local terminologies rather than a quadratic number of mappings.…”

“…Each local schema will be mapped to one ontology in a family of ontologies. For example, we generated a knowledge base of library metadata by mapping from a standard library format, MARC (Machine Readable Cataloging), to an ontology that models bibliographic relationships between works (Weinstein and Birmingham 1998). We were able to specify the mappings in a straightforward way by the simple expedient of permitting the ontology to resemble MARC somewhat more than it should.…”