Fuzzy-Based Methods in Data Analysis with the Focus on Dimensionality Reduction
- 6 Downloads
We analyze a space with a fuzzy partition and show how it determines a measure of closeness. In the space with closeness, we characterize the corresponding Laplace operator and its eigenvectors. The latter serve as projection vectors to reduce the dimension of the original space. We show that the F-transform technique can be naturally explained in the language of dimensionality reduction.
KeywordsFuzzy partition Laplace operator F-transform Dimensionality reduction
This work was supported by the project LQ1602 IT4Innovations excellence in science. The additional support was also provided by the project AI-Met4AI, CZ.02.1.01/0.0/0.0/17-049/0008414.
- 2.J. Janeček, I. Perfilieva, Dimensionality reduction and its F-transform representation, in Proceedings of 11th Conference on EUSFLAT 2019. Atlantis Studies in Uncertainty Modeling (2019)Google Scholar
- 5.C.O.S. Sorzano, J. Vargas, A. Pascual Montano, A survey of dimensionality reduction techniques (2014). arXiv:1403.2877v1 [stat.ML]
- 6.I. Perfilieva, Dimensionality reduction by fuzzy transforms with applications to mathematical finance, in Econometrics for Financial Applications, ed. by L. Anh, L. Dong, V. Kreinovich, N. Thach. ECONVN 2018. Studies in Computational Intelligence (Springer, Cham, 2018), p. 760Google Scholar
- 7.F.R.K. Chung, Spectral Graph Theory (American Mathematical Society, Providence, RI, 1997)Google Scholar