It looks like you're new here. If you want to get involved, click one of these buttons!

- All Categories 1.6K
- General 40
- Azimuth Project 95
- - Latest Changes 345
- News and Information 222
- - Strategy 93
- - Questions 39
- Azimuth Forum 25
- - Conventions and Policies 21
- Chat 166
- Azimuth Blog 133
- - - Action 13
- - - Biodiversity 7
- - - Books 1
- - - Carbon 7
- - - Climate 41
- - - Computational methods 34
- - - Earth science 21
- - - Ecology 40
- - - Energy 28
- - - Geoengineering 0
- - - Mathematical methods 62
- - - Oceans 4
- - - Methodology 16
- - - Organizations 33
- - - People 6
- - - Reports 3
- - - Software 19
- - - Statistical methods 1
- - - Things to do 1
- - - Visualisation 1
- - - Meta 7
- - - Natural resources 4
- Azimuth Wiki 6
- - - Experiments 23
- - - Sustainability 4
- - - Publishing 3
- Azimuth Code Project 69
- - Spam 1

Options

Clustering algorithm on surface temps volumetric grid:

Download video:

Cluster Surface Temps 2010 DOWNLOAD

7 clusters, Euclidean metric (we could use many other metrics).

John this is an example how we could apply machine learning algorithms to volumetric data. We could run many algorithms and track their results graphically to understand some behaviour of the weather system.

Paul Note the Himalays region got clustered rather separately!!!!

Daniel some of your patterns are differentiating boundaries between two clusters, so IMHO they are fundamental to data, I cannot say they are artefacts of any rendering or scaling.

Dara

## Comments

Graham had some R code that computed correlation between segments of data, and he was taking some sort of averages to reduce dimensionality of the grid, I do not think we need to do that. So he could now think of correlation as a METRIC and cluster or computer Nearest Neighbour or compute Similarity and do so without any averaging at all.

In deed following that programming idea he had, we could apply many other metrics as opposed to just the Correlation. So as a matter of fact most statistical operations could be thought of either as metrics of specific kind, or smoothing functions of specific kind.

Dara