tag:blogger.com,1999:blog-6569681.post4056071763451615037..comments2024-01-15T13:17:33.771-08:00Comments on Geeking with Greg: Learning cluster labels from search logsGreg Lindenhttp://www.blogger.com/profile/09216403000599463072noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-6569681.post-933509838606783582008-12-11T23:40:00.000-08:002008-12-11T23:40:00.000-08:00I am interested in this approach to clustering sea...I am interested in this approach to clustering search results too! leveraging search log to help clustering snippets may be the best way for now.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-362948632879244812008-06-15T18:00:00.000-07:002008-06-15T18:00:00.000-07:00Hey, this is a very alegant approach you have sugg...Hey, this is a very alegant approach you have suggested!<BR/>But clusters would depend on amount of data in the logs and I am sure there would be never enough to create a good semantic cluster index. I think your approach is good for finding high frequently used queryies and creating a cluster indexes around them.Unknownhttps://www.blogger.com/profile/05727659920944011056noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-35425648190760332082007-07-04T21:33:00.000-07:002007-07-04T21:33:00.000-07:00This is a nice idea. Indeed, I have used these lab...This is a nice idea. Indeed, I have used these labels to tag my papers in my <A HREF="http://www.christiandelrosso.org/research.php" REL="nofollow"><BR/>home page</A> for already some time and it is quite interesting to see what people search, and at the end, what the search engine provides.cdelrossohttps://www.blogger.com/profile/14725303988710914657noreply@blogger.com