If you are using more than one index then dedup will not work across
indexes. A single index should dedup correctly unless the pages are not
exact duplicates but near duplicates. The dedup process works on url
and byte hash. If the content is even 1 byte different, it doesn't work.
Near duplicate detection is another set of algorithms that haven't been
implemented in Nutch yet. On the query site you can set hte hitsPerSite
to 1 and it should limit your search results.
Dennis
Edward Quick wrote:
Hi,
Eventhough I ran nutch dedup on my index, I still have pages with different
urls but the exactly the same content (see search result example below). From
what I read up on dedup this shouldn't happen though as it deletes the url with
the lowest score. Is there anything else I can try to get rid of these?
Thanks,
Ed.
Item Document :- Client - TeraTerm Pro
... Item Document :- Client - TeraTerm Pro Intranet - Technical Standards
Online Employee Self Service ESS Home ... Description Document
Technology Category: Client Name of item: TeraTerm Pro Related policy: Unix
Access Tool Vendor: Current Technical Status ... standard Telnet tool. Where
printing or keymapping is an issue, TeraTerm ...
http://www.somedomain.com/im/tech/technica.nsf/8918e269a19be23f802563ef004e8e7a/441cdf92bbe06a9e80256c87003d81d9?OpenDocument
(cached) (explain) (anchors)
Item Document :- Client - TeraTerm Pro
... Item Document :- Client - TeraTerm Pro Intranet - Technical Standards
Online Employee Self Service ESS Home ... Description Document
Technology Category: Client Name of item: TeraTerm Pro Related policy: Unix
Access Tool Vendor: Current Technical Status ... standard Telnet tool. Where
printing or keymapping is an issue, TeraTerm ...
http://www.somedomain.com/im/tech/technica.nsf/dacff06c3e1dbc9780257273004e1e3b/441cdf92bbe06a9e80256c87003d81d9?OpenDocument (cached) (explain) (anchors)
_________________________________________________________________
Make a mini you and download it into Windows Live Messenger
http://clk.atdmt.com/UKM/go/111354029/direct/01/
