Skip to main content

Table 2 Examples of SWRL rules in the developing OntoStar

From: Harnessing ontology and machine learning for RSO classification

Background rules

Feature computational rules

so:SpaceObject(?S), so:shape(?S,?Shape), Cynlinder(?Shape), so:height(?S,?H), so:bottom_area(?S,?A), swrlb:multiply(?V,?A,?H) → so:volume(?S,?V)

annotations:

 Source = Mathematics

so:rcs(?X,?RCS), so:power(?P,?D,0.5f), so:SpaceObject(?X), swrlb:divide(?D,?RCS,0.79f), swrlb:subtract(?S,?P,2.57E-13f) → so:size(?X, ?S)

annotations:

 Source = Experience

Semantic classification rule

so:SpaceObject(?S), so:power(?S,?pv), swrlb:greaterThan(?pv,1.0f) → so:Satellite(?S)

annotations:

 Source = Domain

Unordered machine learning rules

Rule from C4.5

so:Satellite(?S), so:MEO(?O), so:inOrbit(?S,?O), so:launchSite(?S,’Cape_Canaveral’) → so:Navigation_Satellite(?S)

annotations:

 Source = mining from UCS_Satellite using J48

 Confidence = 0.9873

 Coverage = 26

so:SpaceObject(?S), so:size(?S,?Sz), swrlb:lessThanOrEqual(?Sz,0.39), so:amr(?S,?A), swrlb: lessThanOrEqual(?A,0.01), so:inOrbit(?S,?O), so:eccentricity(?O,?E), swrlb:greaterThan(?E,0.001426), so:apogee(?O,?AP), swrlb:greaterThan(?AP,784), period(?O,?P), swrlb:greaterThan(?P,108) → so:Debris(?S)

annotations:

 Source = mining from Norad_Catalog using J48

 Confidence = 1

 Coverage = 37