skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Method using a density field for locating related items for data mining

Abstract

A method for locating related items in a geometric space transforms relationships among items to geometric locations. The method locates items in the geometric space so that the distance between items corresponds to the degree of relatedness. The method facilitates communication of the structure of the relationships among the items. The method makes use of numeric values as a measure of similarity between each pairing of items. The items are given initial coordinates in the space. An energy is then determined for each item from the item's distance and similarity to other items, and from the density of items assigned coordinates near the item. The distance and similarity component can act to draw items with high similarities close together, while the density component can act to force all items apart. If a terminal condition is not yet reached, then new coordinates can be determined for one or more items, and the energy determination repeated. The iteration can terminate, for example, when the total energy reaches a threshold, when each item's energy is below a threshold, after a certain amount of time or iterations.

Inventors:
 [1]
  1. Albuquerque, NM
Publication Date:
Research Org.:
Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA (United States)
OSTI Identifier:
874619
Patent Number(s):
US 6424965
Assignee:
Sandia Corporation (Albuquerque, NM)
DOE Contract Number:  
AC04-94AL85000
Resource Type:
Patent
Country of Publication:
United States
Language:
English
Subject:
method; density; field; locating; related; items; data; mining; geometric; space; transforms; relationships; locations; locates; distance; corresponds; degree; relatedness; facilitates; communication; structure; makes; numeric; values; measure; similarity; pairing; initial; coordinates; energy; determined; item; assigned; near; component; draw; similarities; close; force; apart; terminal; condition; reached; determination; repeated; iteration; terminate; example; total; reaches; threshold; below; amount; time; iterations; /707/345/

Citation Formats

Wylie, Brian N. Method using a density field for locating related items for data mining. United States: N. p., 2002. Web.
Wylie, Brian N. Method using a density field for locating related items for data mining. United States.
Wylie, Brian N. 2002. "Method using a density field for locating related items for data mining". United States. https://www.osti.gov/servlets/purl/874619.
@article{osti_874619,
title = {Method using a density field for locating related items for data mining},
author = {Wylie, Brian N},
abstractNote = {A method for locating related items in a geometric space transforms relationships among items to geometric locations. The method locates items in the geometric space so that the distance between items corresponds to the degree of relatedness. The method facilitates communication of the structure of the relationships among the items. The method makes use of numeric values as a measure of similarity between each pairing of items. The items are given initial coordinates in the space. An energy is then determined for each item from the item's distance and similarity to other items, and from the density of items assigned coordinates near the item. The distance and similarity component can act to draw items with high similarities close together, while the density component can act to force all items apart. If a terminal condition is not yet reached, then new coordinates can be determined for one or more items, and the energy determination repeated. The iteration can terminate, for example, when the total energy reaches a threshold, when each item's energy is below a threshold, after a certain amount of time or iterations.},
doi = {},
url = {https://www.osti.gov/biblio/874619}, journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Jan 01 00:00:00 EST 2002},
month = {Tue Jan 01 00:00:00 EST 2002}
}