Automatically Learning Topical Content
The Logos Controlled Vocabulary (LCV) is a controlled vocabulary of some 35,000 concepts from the domain of biblical studies, aligned by hand with a corpus of more than 450,000 articles drawn from hundreds of Bible dictionaries, encyclopedias, glossaries, and journals. This talk describes the LCV and its corpus, why topical content matters for exploring and discovering texts, and the technical challenges involved — including the automatic discovery of topical connections between concepts and articles.