PCA

Edit Answers
1234567891011121314
Across
  1. 3. The process of transforming raw data into a reduced set of representative features.
  2. 6. A direction in the feature space along which the data varies.
  3. 8. The geometric relationship where principal components are at right angles to each other.
  4. 9. The matrix factorization method often used to implement PCA numerically.
  5. 11. A key hyperparameter in t-SNE that balances local and global aspects of the data.
  6. 13. The measure of how much two random variables change together.
  7. 14. The spread of data points that PCA aims to maximize in its first components.
Down
  1. 1. The preprocessing step of scaling data to have zero mean and unit variance.
  2. 2. A scalar representing the amount of variance captured by its associated vector.
  3. 4. The number of input variables or features in a dataset.
  4. 5. A nonlinear technique often used for visualizing high-dimensional clusters.
  5. 7. The phenomenon where data becomes sparse as the number of features increases.
  6. 9. A graphical tool used to determine the number of components to retain.
  7. 10. The process of mapping high-dimensional data onto a lower-dimensional subspace.
  8. 12. A topological space that locally resembles Euclidean space, often targeted by t-SNE.