Ontology Construction Based on Latent Topic Extraction in a Digital Library

摘要

This paper discusses the automatic ontology construction process in a digital library. Traditional automatic ontology construction uses hierarchical clustering to group similar terms, and the result hierarchy is usually not satisfactory for human's recognition. Human-provided knowledge network presents strong semantic features, but this generation process is both labor-intensive and inconsistent under large scale scenario. The method proposed in this paper combines the statistical correction and latent topic extraction of textual data in a digital library, which produces a semantic-oriented and OWL-based ontology. The experimental document collection used here is the Chinese Recorder, which served as a link between the various missions that were part of the rise and heyday of the Western effort to Christianize the Far East. The ontology construction process is described and a final ontology in OWL format is shown in our result.