tag:blogger.com,1999:blog-6569681.post3924110659924556530..comments2024-01-15T13:17:33.771-08:00Comments on Geeking with Greg: Size matters? Or simplicity?Greg Lindenhttp://www.blogger.com/profile/09216403000599463072noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-6569681.post-40890915950711751612008-05-08T07:18:00.000-07:002008-05-08T07:18:00.000-07:00To amend my last comment: I missed the last paragr...To amend my last comment: I missed the last paragraph; the authors tried mixing techniques to reach 98% accuracy.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-27876274835424685902008-05-08T06:33:00.000-07:002008-05-08T06:33:00.000-07:00This doesn't mean the other metrics are useless.I ...This doesn't mean the other metrics are useless.<BR/>I suppose other metrics make different classification errors that the 2000-words predictor, and I expect a combination of several methods would reach a higher score.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-28405092201274533872008-05-04T14:07:00.000-07:002008-05-04T14:07:00.000-07:00We found similar correlations between word count a...We found similar correlations between word count and quality at Epinions. Eric.Eric Goldmanhttps://www.blogger.com/profile/01593907559974631866noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-16484854016344093402008-05-04T14:03:00.000-07:002008-05-04T14:03:00.000-07:00Reminds me of "Pivoted Document Length Normalizati...Reminds me of "Pivoted Document Length Normalization", http://citeseer.ist.psu.edu/singhal96pivoted.html and I guess the avg length of the featured articles could be a parameter into this technique.<BR/><BR/>I never noticed the wikipedia featured articles, but it's great to have that as it's a fun task to try to train a classifier to recognize other "good" articles, and it looks like from the references this has already been explored.<BR/><BR/>-- DaveDavehttps://www.blogger.com/profile/11976158640144983559noreply@blogger.com