Automatic Classification

Classification in Wiki via automatic schemes ThinkingOutLoud.DonaldNoyes.200908252059
In a wiki system, the grouping of concepts and subjects into a document and then giving it a WikiWord name, making its name a DocumentRepresentative, of great use in the Mapping of Concepts. This representation is the first level of grouping wikis use to organize concepts, subjects, concerns, or problems. Because WikiWords are are rarely used in general discourse to identify the document for which it stands, the traditional rigorous techniques for classification will usually ignore them as insignificant. The distribution curve of words which allow WikiWords, will place them on the lower portion of a hyperbolic curve. Some of the automatic classification schemes simply exclude this area.

The referenced Figure demonstrates a hyperbolic curve and how high and low frequency words are filtered out:
WikiStems?, Words, Phrases (Spacified WikiWords)

Some relevance to the inclusion of such things as word stems and phrases in a classification scheme have beeen recognized: "It could equally well be applied to stems of words (or phrases) and in fact this has often been done."
There is a sense in which the use of WikiWords can enhance the location efficiency of searches made with the popular search engines, in that when used as a search term, will return tens to hundreds of results, where using a single word will result in hundred thousands to several million results.

Using WikiWords can be like using a search engine as a powerful magnet to locate the "needle in the haystack".
Other ways the WikiWay can be used to enhance AutomaticClassification:

OneLineSummarizations? Clusterings Categorizations
It might be a great idea, but I suggest that it not create wiki-words automatically. It's best to first be vetted by people.


View edit of September 13, 2010 or FindPage with title or text search