 |
|
| FriendFeed is the easiest way to share online. Learn more » |
Andrew Lang to Andrew's feed, Open Notebook Science Solubility Project
Solvents clustered by Abraham solvation parameters using happieclust. ref: http://friendfeed.com/danielm...
November 14
- Comment
- Like
- Share
Khalid Mirza and Jean-Claude Bradley liked this
This should help us pick sets of diverse solvents.
- Andrew Lang
Is there a reason to use approximate clustering over an exact method? The dataset isn't very large if I recall. Is there a big difference with say standard hierarchical clustering methods?
- Rajarshi Guha
from iPhone
For small datasets it should do exact. Just to make sure I used the the flag --all-pairwise. Didn't seem to change but you're right - I needed to make sure.
- Andrew Lang