Skip to main content
. 2015 May 22;10(5):e0127113. doi: 10.1371/journal.pone.0127113

Table 5. The paired t-test results on the three algorithms with the Null hypothesis ℋ0: the predicted energy values from the two algorithms being considered come from the same distribution.

Algorithms 1D 6D 7D
Result p-value Result p-value Result p-value
GPR vs KNNR Failed to reject ℋ0 0.72 Failed to reject ℋ0 0.94 Failed to reject ℋ0 0.34
GPR vs LINR Rejected ℋ0 7.7 × 10−8 Rejected ℋ0 2.4 × 10−5 Rejected ℋ0 0.01
KNNR vs LINR Failed to reject ℋ0 0.08 Rejected ℋ0 1.9 × 10−2 Failed to reject ℋ0 0.42