Jump to content

Prototype methods

From Wikipedia, the free encyclopedia

Prototype methods are machine learning methods that use data prototypes.[1] A data prototype is a data value that reflects other values in its class,[2] e.g., the centroid in a K-means clustering problem.

Methods

[edit]

The following are some prototype methods[3]

[edit]

While K-nearest neighbor's does not use prototypes, it is similar to prototype methods like K-means clustering.[4]

References

[edit]
  1. ^ Hastie, Trevor. The elements of statistical learning : data mining, inference, and prediction. Tibshirani, Robert,, Friedman, J. H. (Jerome H.) (Second ed.). New York. p. 459. ISBN 9780387848570. OCLC 300478243.
  2. ^ Molnar, Christoph. 6.3 Prototypes and Criticisms | Interpretable Machine Learning.
  3. ^ Hastie, Trevor. The elements of statistical learning : data mining, inference, and prediction. Tibshirani, Robert,, Friedman, J. H. (Jerome H.) (Second ed.). New York. pp. 459–463. ISBN 9780387848570. OCLC 300478243.
  4. ^ Hastie, Trevor. The elements of statistical learning : data mining, inference, and prediction. Tibshirani, Robert,, Friedman, J. H. (Jerome H.) (Second ed.). New York. p. 465. ISBN 9780387848570. OCLC 300478243.