Definitely worth a read   (2011-05-08)
If you have any interest at all in information science, this article is a good one to read. It's not a research report; instead, it is an essay about some of the hottest contemporary issues in information science. I learned a lot by reading it.
What makes this article good is that it offers clear definitions of many of the most important current issues in contemporary information science. It deals a lot with the foundations of the Semantic Web and with some of the most basic problems in information science such as synonyms and homonyms.
Because the abstract already summarizes the article, I won't do that also. However, I will include below several of my favorite quotations from the article.
- "A knowledge organization system's goal in information practice is to support the retrieval process" (p. 1969).
- "An economical principle instructs us not to admit too many concepts into a KOS. If two concepts are more or less similar in terms of extension and intension, these will be regarded as one single “quasi-synonymous” concept" (p. 1954).
- "No concept (and no KOS) remains stable over time" (p. 1955).
- "The incorporation of hyperonyms into a search argument in particular can expand the search results enormously and thus negatively affect precision" (p. 1959).
- "One of the knowledge domains that has enjoyed particular attention in the context of ontological knowledge representation is biology ... But: This [Gene Ontology] is not an ontology at all, it is a thesaurus (to be precise: three partial thesauri for biological processes, molecular functions, and cellular components), as this concept system only uses the PART OF and IS A relations, i.e., only meronymy and hyponymy" (p. 1963).
- "Equally open, in our view, is the question of indexing documents in the Semantic Web. Who will perform this work—the author, the users, or—as automatic indexing—a system?" (p. 1966).
Was this review helpful to you?