Combining bibliometrics, information retrieval, and relevance theory, Part 2: Some implications for information science: Research Articles

摘要: When bibliometric data are converted to term frequency (tf) and inverse document (idf) values, plotted as pennant diagrams, interpreted according Sperber Wilson's relevance theory (RT), the results evoke major variables of information science (IS). These include topicality, in sense intercohesion intercoherence among texts; cognitive effects texts response people's questions; levels expertise a precondition for effects; processing effort textual or other messages received; specificity terms it affects effort; relevance, defined RT effectsseffort ratio; authority their authors. While such concerns figure automatically dialogues between people, they become problematic when people create use judge literature-based systems. The difficulty achieving worthwhile acceptable human-system explains why is central concern IS. Moreover, since relevant communication with both systems unfamiliar uncertain, speakers tend seek that cost them least effort. Yet hearers need greater effort, often specificity, from if responses be highly turn. This theme mismatch manifests itself vague reference questions, underdeveloped online searches, uncreative judging retrieval evaluation trials, perfunctory indexing. Another effect bias toward topical over kinds. can explain these outcomes well more adaptive ones. Pennant applied here literature search Bradford-style journal analysis, model them. Given right context, bibliometrics may predict psychometrics. © 2007 Wiley Periodicals, Inc.