1. We can't find patterns in full attribute space, and patterns may only be found in smaller subspaces.
2. Pattern Trails is an interactive visual approach for the exploration of subspaces of multivariate data
1. Multivariate data analysis and visualization
parallel coordinate plots; pixel bar charts; Chernoff faces.
2. using dimensionality reduction for visual analysis
MDS: multidimensional scaling;
PCA: principal component analysis;
t-SNE: t-distributed stochastic neighbor embedding;
SOM: self-organizing maps
3. Subspace search and visualization
1. Derive interesting subspaces of the multivariate data
1. what are subspace patterns and subspace analysis.
2. The main task in understanding such multivariate data is to identify and interpret relevant patterns like dense groups (clusters), outliers, or correlations.
3. Techniques for visually exploring multivariate data:
Parallel Coordinate Plots;
dimensionality reduction降维; to transform the data to a lower-dimensional space but preserving the main structure of the data.
4. Chernoff Faces