Information Technology Reference
In-Depth Information
Since the matrix obtained is not transitive, its transitive closure is determined
and by that the similarity relation
R
is revealed. Using the decomposition
theorem,
are decomposed to equivalence relations and the obtained results
presented below.
R
Fuzzy clusterization of expert results versus similarity relation
R
Confidence
level
Cluster
{1,2,3,4,5,6}
0,61
0,75
{1,3,6}, {2,4,5}
0,79
{1,6}, {2,4,5},{3}
0,84
{1,6}, {2,4,},{3},{5}
1
{1,6}, {2}, {3}, {4}, {5}
Apparently, 1st and 6th expert results completely coincide, 2nd, 4th and 5th
expert results are similar, but they differ from the results of 1st and 6th experts.
Table 3.8
Parameters of membership functions of term-sets of the formalized expert
approaches
Membership
function
X
X
X
3
1
2
μ
(0;0,075;0;0,05)
(0;0,05;0;0,1)
(0;0,025;0;0,05)
i
μ
(0,125;0,05;0,05)
(0,15;0,1;0,1)
(0,075;0,05;0,05)
i
2
μ
(0,175;0,225;0,05;0,15)
(0,25;0,3;0,1;0,2)
(0,125;0,35;0,05;0,1)
i
3
μ
(0,375;0,425;0,15;0,15)
(0,5;0,2;0,2)
(0,45;0,1;0,1)
i
4
μ
(0,575;0,15;0,15)
(0,7;0,775;0,2;0,15)
(0,55;0,6;0,1;0,2)
i
5
μ
(0,725;1;0,15;0)
(0,925;1;0,15;0)
(0,8;1;0,2;0)
i
6
X
X
X
4
5
6
(0;0,05;0;0,1)
(0;0,025;0;0,05)
(0;0,075;0;0,05)
(0,15;0,175;0,1;0,15)
(0,075;0,175;0,05;0,15)
(0,125;0,05;0,05)
(0,325;0,375;0,15;0,15)
(0,325;0,15;0,15)
(0,175;0,225;0,05;0,15)
(0,525;0,15;0,15)
(0,475;0,15;0,15)
(0,375;0,425;0,15;0,15)
(0,675;0,7;0,15;0,2)
(0,625;0,85;0,15;0,1)
(0,575;0,15;0,15)
(0,9;1;0,2;0)
(0,95;1;0,1;0)
(0,725;1;0,15;0)
