Table 5 Optimized health prediction model for WSN.

From: A new health prediction model for a sensor network based on belief rule base with attribute reliability

No

Rule weight

Feature

Output distribution

\(AR\)

\(FR\)

\(\{ H,\;M,\;L\}\)

1

0.4857

\(L\)

\(L\)

(0.9248 0.0072 0.0680)

2

0.2223

\(L\)

\(SL\)

(0.9293 0.0571 0.0136)

3

0.0030

\(L\)

\(M\)

(0.3626 0.0091 0.6283)

4

0.0019

\(L\)

\(SH\)

(0.8781 0.1161 0.0057)

5

0.3637

\(L\)

\(H\)

(0.0000 0.1918 0.8085)

6

0.0000

\(M\)

\(L\)

(0.1931 0.8066 0.0003)

7

0.2912

\(M\)

\(SL\)

(0.9985 0.0025 0.0000)

8

0.9306

\(M\)

\(M\)

(0.9992 0.0011 0.0000)

9

0.6959

\(M\)

\(SH\)

(0.9963 0.0039 0.0000)

10

0.5965

\(M\)

\(H\)

(0.9952 0.0036 0.0012)

11

0.9973

\(SH\)

\(L\)

(0.0003 0.0014 0.9983)

12

0.0000

\(SH\)

\(SL\)

(0.0014 0.0016 0.9970)

13

0.3308

\(SH\)

\(M\)

(0.0053 0.0322 0.9625)

14

0.0137

\(SH\)

\(SH\)

(0.5073 0.4924 0.0003)

15

0.6357

\(SH\)

\(H\)

(0.1238 0.3275 0.5487)

16

0.0036

\(H\)

\(L\)

(0.0721 0.2685 0.6594)

17

0.8106

\(H\)

\(SL\)

(0.0016 0.0018 0.9966)

18

0.8338

\(H\)

\(M\)

(0.0000 0.0042 0.9964)

19

0.9966

\(H\)

\(SH\)

(0.0033 0.0012 0.9954)

20

0.0000

\(H\)

\(H\)

(0.0384 0.5554 0.4062)