Welcome to WebmasterWorld Guest from 188.8.131.52
Forum Moderators: open
Iam a little bit confused about Beta FAST Topics:
A clients site there is no title on it and there is no odp listing and the site is not crawled after i have done a SEO
But when i search for the kw`s i got a first place in Beta FAST Topics: Where comes the results from?
i think its strange but there is a lot of thing that is strange
Use this set (odp cats)as a training set for classification, i.e. based on the documents in the existing hierarchy, add more documents from the web using document similarity between these and those already classified.
Generate new groups (clusters) for those new documents that cannot be matched properly and use key terms from the documents within this group to label it.
they could then run a "similarity" algorithm across sites in each category, to find other sites in the FAST index which are similar to those sites in any given category, but not already included in that category via the ODP. then they'd add those sites to their version of the category.
thing to remember is that these calculations must be done in advance, with the categories cached. there's no way they'd be able to run the calculations on the fly, across the entire database.
you'd think they'd go ahead and crawl ODP, but who knows with the PFI stuff going on. maybe they only want to crawl submitted stuff...
anyway, our stuff that is in FAST but not in ODP seems to be categorized pretty well; and you seem to be able to find the pages in multiple categories too...