Skip to main content

Table 4 The characteristics of rule sets generated by LR-Rules: the value of the integrated Brier score (IBS), the number of generated rules (#rules), the average rule length, the average rule coverage (%cov), a percentage of significant rules (p-value of log-rank test with FDR adjustment below 0.05; %sign)

From: Learning rule sets from survival data

Dataset IBS #rules Length %cov %sign
actg320 0.0597 13.1 3.7 28 99.18
BMT-Ch 0.2231 4.3 3.3 48 98.40
cancer 0.1538 7.3 3.4 40 97.35
follic 0.1896 5.0 1.8 38 99.32
GBSG2 0.1783 7.7 2.7 36 99.75
hd 0.2185 4.6 1.3 44 99.43
LAC 0.2116 2.8 6.8 41 100.00
lung 0.1472 5.1 1.0 45 100.00
Melanoma 0.1823 6.2 2.6 33 99.69
mgus 0.1719 7.4 3.2 34 99.70
PTC 0.1426 18.7 3.0 22 79.69
pbc 0.1502 3.7 1.1 45 100.00
std 0.2220 17.8 6.5 25 98.74
uis 0.1539 5.5 1.2 41 100.00
wcgs 0.0432 21.8 5.2 19 98.95
whas1 0.2130 3.8 1.3 43 100.00
whas500 0.2046 4.6 1.2 40 100.00
zinc 0.0931 7.1 2.5 31 98.42