COPYRIGHTS

This package includes the externally sourced Vowel Recognition dataset and
synthetic Gaussian cluster datasets generated by the package authors.
The relevant copyright and license notices are summarized below.

1. Vowel Recognition dataset

Distributed dataset:
  - vowel (data/vowel.rda)

Source:
  UCI Machine Learning Repository, Vowel Recognition (Deterding Data).
  https://doi.org/10.24432/C58P4S

Original dataset creators:
  David Deterding, Mahesan Niranjan, and Tony Robinson.

Context-enriched version:
  The included data were prepared from vowel-context.data, which adds
  explicit speaker information, by Peter Turney. See Turney (1993),
  "Robust Classification With Context-Sensitive Features."

Use in this package:
  The dataset has 990 observations and 14 variables. The categorical
  codes for the training/test split, speaker, sex, and vowel class were
  converted to factors; the ten numeric acoustic features were retained
  without alteration.

License:
  Creative Commons Attribution 4.0 International (CC BY 4.0).
  https://creativecommons.org/licenses/by/4.0/

Attribution:
  Deterding, D., Niranjan, M., and Robinson, T. (1988).
  Connectionist Bench (Vowel Recognition - Deterding Data).
  UCI Machine Learning Repository.

  Turney, P. (1993).
  Robust Classification With Context-Sensitive Features.

License notice:
  CC BY 4.0 permits sharing and adaptation subject to attribution,
  provision of a license link, and indication of changes. For full
  attribution and transformation details, see the package LICENSE.note.

2. Synthetic Gaussian cluster datasets

Datasets:
  - gau
  - gau_g

Source:
  Generated by the authors of this package.

Use in this package:
  The dataset `gau` contains synthetic 20-dimensional Gaussian cluster
  observations for reproducible examples of PCA and differentially
  private PCA visualization. The dataset `gau_g` contains the same
  observations and numeric features with an additional group label.

License:
  These synthetic datasets are distributed under the same MIT license
  as the package code.