Creation of Ontological Knowledge Bases in the Semantic Web …
225
An analysis of expression (29) taking into account expressions (22–28) shows that
the parameter has the greatest impact on the complexity of the proposed algorithmic
support by the criterion r of time costs, and the increase in the complexity of OKB
formation with an increase in the number of objects has a polynomial character.
For experimental verification of the effectiveness of the algorithmic support
proposed by the author in comparison with other implementations, we consider
the implementation of the algorithmic support for the formation of OKB based on
the widely used StringSearch object search method described in WebProtege. The
StringSearch method searches for objects by the symbolic names of ODR objects
without taking into account their structural and logical characteristics. Thus, UECC
of forming OKB (T max
wb ) based on StringSearch, by analogy with the algorithmic
software developed by the author, is represented as the time spent on processing
r—ODR objects from which OKB is formed. A feature of the implementation of this
algorithmic support is that when adding the k-th ODR object to OKB, it is necessary
to loop through at most k-1 OKB objects and add this (k-th) ODR object to it with a
superclass object or subclass object. The mathematical expression of the UECC of
this algorithmic support is presented below:
T max
wb (r) =
r
m=1
(m − 1)ct
cp
s + ct
+
o + ct
+
d + ct
+
f + ct
+
i .
(31)
Let us compare experimentally the above implementations of the algorithmic
support for the formation of OKB.
The experimental values of the UECC OKB formation depend on the software
implementation and configuration of computer technology. Table 1 presents the
average values of time spent on operations.
The maximum values of the time costs of operations (22–28) are presented below:
ct
cp
s = 118.619(mcs)
Table 1 Time spent on operations (22–28), MCS
ct
cp
s
ct
cp
d
ct
cp
f
ct +
o
ct
+
d
ct
+
f
ct
+
i
106.426
561.46
354.474
1234.37
638.56
667.89
75,713.3
110.346
582.41
331.848
1248.62
663.696
693.03
66,369.6
106.256
563.56
333.365
1264.43
638.556
687.12
63,101.4
118.619
582.44
316.764
1238.69
643.597
673.43
63,855.6
106.726
561.46
332.344
1280.46
648.559
657.39
64,903.1
110.686
583.47
317.654
1249.12
669.611
697.86
64,484.1
106.425
561.46
324.798
1235.41
632.665
687.59
64,903.1
110.116
582.41
306.708
1252.32
683.611
663.07
64,884.8
106.436
561.46
331.12
1272.92
631.536
657.49
64,400.3
110.623
582.69
316.764
1276.27
661.196
643.08
63,384.1
225
An analysis of expression (29) taking into account expressions (22–28) shows that
the parameter has the greatest impact on the complexity of the proposed algorithmic
support by the criterion r of time costs, and the increase in the complexity of OKB
formation with an increase in the number of objects has a polynomial character.
For experimental verification of the effectiveness of the algorithmic support
proposed by the author in comparison with other implementations, we consider
the implementation of the algorithmic support for the formation of OKB based on
the widely used StringSearch object search method described in WebProtege. The
StringSearch method searches for objects by the symbolic names of ODR objects
without taking into account their structural and logical characteristics. Thus, UECC
of forming OKB (T max
wb ) based on StringSearch, by analogy with the algorithmic
software developed by the author, is represented as the time spent on processing
r—ODR objects from which OKB is formed. A feature of the implementation of this
algorithmic support is that when adding the k-th ODR object to OKB, it is necessary
to loop through at most k-1 OKB objects and add this (k-th) ODR object to it with a
superclass object or subclass object. The mathematical expression of the UECC of
this algorithmic support is presented below:
T max
wb (r) =
r
m=1
(m − 1)ct
cp
s + ct
+
o + ct
+
d + ct
+
f + ct
+
i .
(31)
Let us compare experimentally the above implementations of the algorithmic
support for the formation of OKB.
The experimental values of the UECC OKB formation depend on the software
implementation and configuration of computer technology. Table 1 presents the
average values of time spent on operations.
The maximum values of the time costs of operations (22–28) are presented below:
ct
cp
s = 118.619(mcs)
Table 1 Time spent on operations (22–28), MCS
ct
cp
s
ct
cp
d
ct
cp
f
ct +
o
ct
+
d
ct
+
f
ct
+
i
106.426
561.46
354.474
1234.37
638.56
667.89
75,713.3
110.346
582.41
331.848
1248.62
663.696
693.03
66,369.6
106.256
563.56
333.365
1264.43
638.556
687.12
63,101.4
118.619
582.44
316.764
1238.69
643.597
673.43
63,855.6
106.726
561.46
332.344
1280.46
648.559
657.39
64,903.1
110.686
583.47
317.654
1249.12
669.611
697.86
64,484.1
106.425
561.46
324.798
1235.41
632.665
687.59
64,903.1
110.116
582.41
306.708
1252.32
683.611
663.07
64,884.8
106.436
561.46
331.12
1272.92
631.536
657.49
64,400.3
110.623
582.69
316.764
1276.27
661.196
643.08
63,384.1