Has anybody written any code to derive a TW from
- a set of postings to a forum
- postings to a mailing list
- a set of HTML documents
It occurs to me that h1 tags could be used to split HTML into tiddlers
and things like TF*IDF algorithms could be used to derive keywords
One would also need to transform HTML (or text) into wiki text