I need help. I don't understand how the "recall" comes out , or how does
Excite know the total of the relevant documents in other services ?
Thanks,
Yaokun
Yaokun@aol.com
>>>>>>>>>>
(http://www.excite.com/ice/precision.html)
Excite has the highest recall and the highest precision of any
search engine currently available. How do we know? Well, we picked a
few
sample queries and performed searches on Excite and on our
competitors.
For each document in the top 20 hits on any service, we judged the
document
to be relevant or nonrelevant. (We used several techniques to
determine
whether or not a page was relevant. We considered a page to be
nonrelevant
if (1) the page was no longer available; (2) the page was a
duplicate of a page
seen earlier in the service's hit list; and (3) the page wasn't
really about the
query topic.)
Then, using this master list of relevant documents, we went back and
calculated the precision and recall for each service using result
lists of various
sizes. Finally, we made a graph for each service based on different
pairs of
precision and recall.
<<<<<<<<<<<
_________________________________________________
This messages was sent by the robots mailing list. To unsubscribe, send mail
to robots-request@webcrawler.com with the word "unsubscribe" in the body.
For more info see http://info.webcrawler.com/mak/projects/robots/robots.html