@article{fdi:010066930,
  title = {{D}eep kernel dimensionality reduction for scalable data integration},
  author = {{S}okolovska, {N}. and {C}l{\'e}ment, {K}. and {Z}ucker, {J}ean-{D}aniel},
  editor = {},
  language = {{ENG}},
  abstract = {{D}imensionality reduction is used to preserve significant properties of data in a low dimensional space. {I}n particular, data representation in a lower dimension is needed in applications, where information comes from multiple high dimensional sources. {D}ata integration, however, is a challenge in itself. {I}n this contribution, we consider a general framework to perform dimensionality reduction taking into account that data are heterogeneous. {W}e propose a novel approach, called {D}eep {K}ernel {D}imensionality {R}eduction which is designed for learning layers of new compact data representations simultaneously. {T}he method can be also used to learn shared representations between modalities. {W}e show by experiments on standard and on real large-scale biomedical data sets that the proposed method embeds data in a new compact meaningful representation, and leads to a lower classification error compared to the state-of-the-art methods.},
  keywords = {{D}imensionality reduction ; {H}eterogeneous data integration},
  booktitle = {},
  journal = {{I}nternational {J}ournal of {A}pproximate {R}easoning},
  volume = {74},
  numero = {},
  pages = {121--132},
  ISSN = {0888-613{X}},
  year = {2016},
  DOI = {10.1016/j.ijar.2016.03.008},
  URL = {https://www.documentation.ird.fr/hor/fdi:010066930},
}