Talk:Document clustering: Difference between revisions

Content deleted Content added
SineBot (talk | contribs)
m Signing comment by 162.138.210.3 - ""
Disputed: new section
Line 5:
== FirstGov.gov example missing clustering? ==
I was unable to find clustering on the provided example link for usa.gov. Maybe [http://new2.yippy.com/search?query=ice+cream&tb=sitesearch-all&v%3Aproject=clusty-new Yippy] would be a better example? <span style="font-size: smaller;" class="autosigned">— Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[Special:Contributions/162.138.210.3|162.138.210.3]] ([[User talk:162.138.210.3|talk]]) 18:15, 18 December 2013 (UTC)</span><!-- Template:Unsigned IP --> <!--Autosigned by SineBot-->
 
== Disputed ==
 
The factual accuracy of this article is quite low. First off, it discusses both ad hoc clustering (in search engines) and what might be called full-set clustering, without acknowledging the difference. Then, it makes a few unintelligible remarks like "Document clustering is generally considered to be a centralized process" (?) and finally it claims that k-means is "less accurate" than hierarchical clustering, as if accuracy on clustering can be measured. (Surely it can when ground truth labels are provided, but I've never seen a comparison that showed hierarchies to be more accurate.) [[User:Qwertyus|Q<small>VVERTYVS</small>]] <small>([[User talk:Qwertyus|hm?]])</small> 13:33, 21 March 2014 (UTC)