Many dimensions. One sky.Explore your data through principal component analysis.
Principal component analysis (PCA) finds the directions in which your data varies most. We center your numeric columns, optionally scale them to equal variance, and project each row onto the first three principal components.
Each row becomes a soft particle in the cloud. Switch to Data points to see the exact projected positions. Density and particle size change the rendering, not the PCA results.
The explained variance tells you how much of the original variation these three dimensions preserve. Non-numeric columns are excluded; rows with missing numeric values are skipped. Large datasets are evenly sampled to 2,500 rows before PCA.
The sample is synthetic atmospheric data, created to demonstrate correlated measurements. The cloud is an artistic rendering, not a weather simulation.