Talk:Concept mining

From Wikipedia, the free encyclopedia

Taxonomic classification is not the same thing as concept mining. Although Concept Mining can be used for taxonomic classification, i.e. looking at a collection of documents, and inferring which taxonomic class they belong to, such taxonomies are normally constructed for a particular task, for instance a particular domain of medicine, whereas concept mining is concerned with generating analyses of greneral purpose documents based on the conceptual relationships in common speech. The uses of this analysis extend far beyond the librarianship applications of taxonomic classification. —The preceding unsigned comment was added by Scientio (talk • contribs) 15:21, 22 November 2006 (UTC).

I agree. Concept mining involves what data you look at when you're trying to understand a document's content (mapping words to concepts instead of using word counts directly) while taxonomic classification involves what you do with your understanding (placing documents into a predetermined tree, rather than grouping them into unhierarchical clusters, ranking them for relevance, or whatever else). The two concepts seem different enough that I see no justification for a merge. —David Eppstein 16:57, 23 November 2006 (UTC)
OK, I see your distinction. But if your talking about placing documents into a predetermined tree I think this is exactly concept mining, or at least Text classification, mapping of documents to concepts, which already exists in some given hierarchy, and ranking them for relevance, is more concept mining than taxonomic classification. From the definition of taxonomic classification, it sounds like it's the act of building the taxonomy. Roee 00:07, 29 November 2006 (UTC)
I thought "concept mining" was an associated concept for computer science? Yes, taxonomic classification is a hierarchical method of arrangement or classification but is a broad term not specific to computer science terminologies with many applications from biology to epistemology ... Stevenmitchell 00:47, 10 December 2006 (UTC)