Cluster analysis
Groups samples by chemical composition. Numeric columns are normalised so that each sample sums to 100 per cent and are then standardised. The result is the input table with an additional cluster column, and a plot of the first two components.
Input
Select a data file
CSV or Excel, one row per sample
Method
K-means is suited to compact groups of similar size.
3
Data preview
No file selected.
Results
No results yet. Select a file and run the analysis.