Linear mappings of local data structures

Mayer Aladjem; Its'hak Dinstein

doi:10.1016/0167-8655(92)90054-4

Linear mappings of local data structures

Mayer Aladjem, Its'hak Dinstein

Department of Electrical & Computer Engineering

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

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 language	English
Pages (from-to)	153-159
Number of pages	7
Journal	Pattern Recognition Letters
Volume	13
Issue number	3
DOIs	https://doi.org/10.1016/0167-8655(92)90054-4
State	Published - 1 Jan 1992

Keywords

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

Access to Document

10.1016/0167-8655(92)90054-4

Cite this

@article{a946c127304d442baefe91fcba7e2eea,

title = "Linear mappings of local data structures",

abstract = "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.",

keywords = "Interactive pattern recognition, cluster analysis, local data structure, mapping of multidimensional data",

author = "Mayer Aladjem and Its'hak Dinstein",

note = "Funding Information: This work was supported in part by the Israeli Ministry of Science and Technology under contract number 3528, and in part by the Paul lvanier Center for Robotics and Production Management, Ben Gurion University of the Negev, Israel.",

year = "1992",

month = jan,

day = "1",

doi = "10.1016/0167-8655(92)90054-4",

language = "English",

volume = "13",

pages = "153--159",

journal = "Pattern Recognition Letters",

issn = "0167-8655",

publisher = "Elsevier",

number = "3",

}

TY - JOUR

T1 - Linear mappings of local data structures

AU - Aladjem, Mayer

AU - Dinstein, Its'hak

N1 - Funding Information: This work was supported in part by the Israeli Ministry of Science and Technology under contract number 3528, and in part by the Paul lvanier Center for Robotics and Production Management, Ben Gurion University of the Negev, Israel.

PY - 1992/1/1

Y1 - 1992/1/1

N2 - 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.

AB - 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.

KW - Interactive pattern recognition

KW - cluster analysis

KW - local data structure

KW - mapping of multidimensional data

UR - http://www.scopus.com/inward/record.url?scp=44049121522&partnerID=8YFLogxK

U2 - 10.1016/0167-8655(92)90054-4

DO - 10.1016/0167-8655(92)90054-4

M3 - Article

AN - SCOPUS:44049121522

SN - 0167-8655

VL - 13

SP - 153

EP - 159

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

IS - 3

ER -

Linear mappings of local data structures

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this