Method using a density field for locating related items for data mining
- Albuquerque, NM
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.
- Research Organization:
- SANDIA CORP
- DOE Contract Number:
- AC04-94AL85000
- Assignee:
- Sandia Corporation (Albuquerque, NM)
- Patent Number(s):
- US 6424965
- OSTI ID:
- 874619
- Country of Publication:
- United States
- Language:
- English
Similar Records
Method of locating related items in a geometric space for data mining
Method of data mining including determining multidimensional coordinates of each item using a predetermined scalar similarity value for each item pair
Related Subjects
amount
apart
assigned
below
close
communication
component
condition
coordinates
corresponds
data
degree
density
determination
determined
distance
draw
energy
example
facilitates
field
force
geometric
initial
item
items
iteration
iterations
locates
locating
locations
makes
measure
method
mining
near
numeric
pairing
reached
reaches
related
relatedness
relationships
repeated
similarities
similarity
space
structure
terminal
terminate
threshold
time
total
transforms
values