PreFace: Faceted Retrieval of Prerequisites Using Domain-Specific Knowledge Bases

Prajna Upadhyay, Maya Ramanath

Abstract: While learning new technical material, a user faces difficulty encountering new concepts for which she does not have the necessary prerequisite knowledge. Determining the right set of prerequisites is challenging because it involves multiple searches on the web. Although a number of techniques have been proposed to retrieve prerequisites, none of them consider grouping prerequisites into interesting facets. To address this issue, we have developed a system called PreFace that (i) automatically determines interesting facets for a given concept of interest, and, (ii) determines prerequisites for the concept and facet. The key component of PreFace is a retrieval model that balances the trade-off between the relevance of the facets and their diversity. We achieve this by representing each facet as a language model estimated using a domain-specific knowledge base and a large corpus of research papers, and ranking them using a risk-minimization framework. Our evaluation of the results over a benchmark set of queries shows that PreFace retrieves better facets and prerequisites than state-of-the-art facet extraction techniques.

External IDs:dblp:conf/semweb/UpadhyayR20