Talk:Consensus clustering

Needs expansion
Please, more reference and expand description of the alghorithms. 00:55, 7 October 2010 178.73.63.95 (talk)
 * Yes, effectively no information here. Melcombe (talk) 16:46, 13 December 2010 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 external links on Consensus clustering. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20081201150950/http://www.ideal.ece.utexas.edu/papers/2007/punera07softconsensus.pdf to http://www.ideal.ece.utexas.edu/papers/2007/punera07softconsensus.pdf
 * Added archive https://web.archive.org/web/20060828084525/http://www.cs.helsinki.fi/u/tsaparas/publications/aggregated-journal.pdf to http://www.cs.helsinki.fi/u/tsaparas/publications/aggregated-journal.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 08:35, 12 August 2017 (UTC)

Problems with the PAC score
I noticed the author of the PAC score paper has inserted a section on their method with graphics on this page. I think this is fine, as it is an important addition to the Monti consensus clustering method (which I have now added mathematical detail on), although I did reduce the detail on this one paper a bit, because I thought it was over glorified. We have found in practice the PAC score works poorly because it overestimates K all the time, in the scientific reports paper the method was not tested on real data, instead just simulations were used. We have fixed this method by adding a Monte Carlo simulation to the Monti consensus clustering algorithm (M3C). I added a brief few sentences to explain this method. However, there is a wikipedia user 'HelpUsStopSpam' who thought this section should be deleted because I am the primary author of the paper. I really do think a new comment should be added at somepoint to explain the inherant bias of the Monti consensus clustering algorithm and the PAC score. A sentence or two is all that is required. Basically, as K increases so does the stability of the consensus matrix on random data, this is a major problem we have noticed again and again on cancer and other datasets. Thank you.