Linear mappings of local data structures

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


Two methods for linear mapping of multidimensional data in the case of unsupervised learning are proposed. The first method maximizes the mean square density gradient of the projected samples with the intention of compressing the clusters. The second method is based on the k-NN technique and obtains a map of the scatter of the neighbor clusters. An experiment with the classical Iris data shows the mapping accuracy of the latter method.

Original languageEnglish
Pages (from-to)153-159
Number of pages7
JournalPattern Recognition Letters
Issue number3
StatePublished - 1 Jan 1992


  • Interactive pattern recognition
  • cluster analysis
  • local data structure
  • mapping of multidimensional data

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence


Dive into the research topics of 'Linear mappings of local data structures'. Together they form a unique fingerprint.

Cite this