Comments on: Quote of the Week, Aug 23, 2007 http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/ Weaving together Astronomy+Statistics+Computer Science+Engineering+Intrumentation, far beyond the growing borders Fri, 01 Jun 2012 18:47:52 +0000 hourly 1 http://wordpress.org/?v=3.4 By: aconnors http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/comment-page-1/#comment-77 aconnors Fri, 24 Aug 2007 20:47:58 +0000 http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/#comment-77 <p>Very interesting quotes, Hyunsook!</p> <p>Speaking of questioning whether classifications and correlations are instrumental versus intrinic to the physics, I note this has recently appeared on astro-ph:<br /> http://adsabs.harvard.edu/abs/2007arXiv0706.1275B</p> <p>I haven't worked through their arguments myself, so I can't speak for it; but the ideas are certainly worth discussing in some depth.</p> Very interesting quotes, Hyunsook!

Speaking of questioning whether classifications and correlations are instrumental versus intrinic to the physics, I note this has recently appeared on astro-ph:
http://adsabs.harvard.edu/abs/2007arXiv0706.1275B

I haven’t worked through their arguments myself, so I can’t speak for it; but the ideas are certainly worth discussing in some depth.

]]> By: hlee http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/comment-page-1/#comment-76 hlee Fri, 24 Aug 2007 17:16:19 +0000 http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/#comment-76 <p>An excerpt from <a href="http://www.amazon.com/Cluster-Analysis-Brian-S-Everitt/dp/0340761199/ref=pd_bbs_1/105-1140178-6406047?ie=UTF8&s=books&qid=1187972525&sr=8-1" rel="nofollow"> Cluster Analysis by Everitt, Landau, and Leese. </a></p> <p><i>It is generally impossible a priori to anticipate what combination of variables, similarity measures and clustering techniques is likely to lead to interesting and informative classification.</i> Consequently, the analysis proceeds through several stages, with the researcher intervening if necessary to alter <strong>variables</strong>, choose a different <strong>similarity measure,</strong> concentrate on a particular <strong>subset of individuals,</strong> and so on. The final, extremely important, stage concerns the <strong>evaluation</strong> of the clustering solutions obtained. <i>Are the clusters real or merely artifacts of the algorithms? Do other solutions exist which are better?<i></p> An excerpt from Cluster Analysis by Everitt, Landau, and Leese.

It is generally impossible a priori to anticipate what combination of variables, similarity measures and clustering techniques is likely to lead to interesting and informative classification. Consequently, the analysis proceeds through several stages, with the researcher intervening if necessary to alter variables, choose a different similarity measure, concentrate on a particular subset of individuals, and so on. The final, extremely important, stage concerns the evaluation of the clustering solutions obtained. Are the clusters real or merely artifacts of the algorithms? Do other solutions exist which are better?

]]> By: hlee http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/comment-page-1/#comment-75 hlee Fri, 24 Aug 2007 16:16:22 +0000 http://hea-www.harvard.edu/AstroStat/slog/2007/quote-of-the-week-aug-23-2007/#comment-75 <p>Hilarious and witty but it tells a lot on clustering. When it comes to clustering, eyes are the best but unfortunately, eyes cannot do much with higher dimensional data. Statistics and machine learning are tools to assist eyes but, as David van Dyk pointed out, choices of dimension reduction or parameter transformation methods make results controversial. What should we follow? scientific expertise or statistical optimality?</p> Hilarious and witty but it tells a lot on clustering. When it comes to clustering, eyes are the best but unfortunately, eyes cannot do much with higher dimensional data. Statistics and machine learning are tools to assist eyes but, as David van Dyk pointed out, choices of dimension reduction or parameter transformation methods make results controversial. What should we follow? scientific expertise or statistical optimality?

]]>