Holistic Graph-Based Representation and AI for Digital Library Management

Abstract

The traditional record-based approach used in Digital Libraries has gone as far as it could. To support the needs and activities of different kinds of users, we propose a graph-based ‘holistic’ representation of DL knowledge, describing the documents’ metadata, physical aspects, content, context and lifecycle. We also propose its implementation based on a framework in which Formal Ontologies play the role of data schemas for the LPG-based DBs. This enables automated reasoning on the DL data, and we propose the use of multiple inference strategies.