Some of the big data proponents are interested more in model-less studies, a/b testing. They say specialist knowledge isn't needed, actually it gets in the way,the data will provide the results (http://www.slate.com/articles/health_and_science/new_scientist/2012/12/kaggle_president_jeremy_howard_amateurs_beat_specialists_in_data_prediction.html).
The following put a good case for models, and how modelless data might be flawed.
http://scensci.wordpress.com/2012/12/14/big-data-or-pig-data/ [Big Data or Pig Data? (A fable on huge amounts of data and why we [don't] need models)
Cheers
Brian
On Tue, Feb 5, 2013 at 2:46 PM, Bjonnh <bjonnh-diybio@bjonnh.net> wrote:
On Mon, Feb 04, 2013 at 03:56:26PM -0700, leaking pen wrote:A model is not devoted to an explication of why it works. But used to
> There's a difference between a setup, "here's what what we did and what
> happened" and a model of why it works.
try to find how you can try to explain it. By making a description
of a phenomenon, you try to mimic the characteristics of this
phenomenon. It's not meant to be true or false, but it needs to be able to be
falsifiable (proven that it's wrong or not accurate). And to be
falsifiable it means that you need to be able to reproduce it and
compare the whole set of results between the experiments…
--
Jonathan BISSON
PhD in Science, Technology and Health - Speciality : Chemistry-Biology interface
Actually in Oenology post-doc until end of February at :
Laboratoire d'Oenologie - Équipe Philippe Darriet
Institut des Sciences de la Vigne et du Vin (ISVV)
210 Chemin de Leysotte CS 50008
F-33882 Villenave d'Ornon Cedex
Cell: +33 (0) 6 81 39 02 92
E-mail (please use this one from now) : research@bjonnh.net
----------------------------------------
Brian Degger
twitter: @drbrian
----------------------------------------
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@googlegroups.com. To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+unsubscribe@googlegroups.com.
To post to this group, send email to diybio@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
0 comments:
Post a Comment