Re: Dynamic taxonomy
| Newsgroups | gmane.comp.infodesign.facetedclassification |
|---|---|
| Message-ID | <[email protected]> |
Hello sorry for the delay and thanks for your comment. Looks like there are lots of Italians in this board. > (2) Although dynamic taxonomy approach uses ad-hoc > categories basically analogous to facets, facets to be > used are not necessarily the canonical ones of object, ... Precisely, I am not yet convinced that canonical facets are a good idea and I feel like sticking to my more general multidimensional classification scheme. The main reasons are basically 2: 1) my framework has a much more general scope than faceted classification: going from e-commerce to diagnostic systems. In this context, canonical facets have little sense, if any. 2) the extensional inference rule, and the paradigm of exploration by subsequent foci changes things quite a bit from traditional facet approaches and may warrant a rethinking of the entire facet approach. > The additional original element in the description > of dynamic taxonomies seems to be the reuse of > a posteriori information about connection between > subjects, on the basis of their actual occurrence in > documents (which would be called the _literary warrant_ > by LIS people): I would be curious to know which kind > of use is then provided for this information. I believe this is the most important departure from traditional approaches. The extensional inference rule allows to summarize any set of documents according to the original taxonomy and effectively guides the user to reach his goal, in a simple, unobtrusive way. You can play with a few demos at www.knowledgeprocessors.com ---------- Thanks for playing. To unsubscribe from this group, send an email to: [email protected] Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/