Transparency and the future of semantic searching in academic libraries

Abstract

“Semantic search” is a fairly broad term without a fixed goal amongst developers. Most agree that semantic searching should reach beyond simple keyword or text/string matching in order to provide more robust taxonomies and relevant information retrieval systems. Many novice users and first-year students struggle to retrieve the scholarly sources they desire because they lack the specialized vocabulary attained through advanced years of study. Most students are used to the dominant semantic search discovery system, Google and Google Scholar, but few students understand how these systems work. Query parsing, fuzzy matching, and understanding how semantic searching utilizes taxonomies for more accurate tagging is not usually a consideration for a desperate sophomore looking for last-minute sources to finish a paper that is due the next day. So, there is a danger that meeting a student’s perceived need for a Google-like discovery system is more important than creating a transparent system. However, information literacy and library instruction must give careful consideration to these issues and be able to help emerging scholars/students understand both the ethical and practical horizons of semantic search tools.