Skip to main content

Advertisement

Table 1 Performance of each algorithm on three synthetic data sets after 500 trials, with the initial number of Gaussian components is set as k =20, where ‘a’ indicates the best within its column

From: Projection-embedded BYY learning algorithm for Gaussian mixture-based clustering

Data set GMM-a GMM-b GMM-c
Algorithms CSR VI PRI CSR VI PRI CSR VI PRI
VB-DNW 0.4660 1.0243 0.7730 0.5160 0.6264 0.8599 0.1060 1.3337 0.6469
MML-Jef 0.1700 3.2637 0.7345 0.1600 4.8235 0.7573 0.4140 58.0039 0.6388
BYY-Jef 0.2167 1.1135 0.7006 0.5533 0.6650 0.8257 0.0100 1.6889 0.4732
BYY-DNW 0.1433 1.1947 0.7039 0.0700 0.5373 a 0.8760 0 1.7948 0.4622
pBYY 0.7260 a 0.5852 a 0.8692 a 0.8840 a 0.5482 0.8779 a 0.6100 a 1.1328 a 0.7451 a
  1. For a good performance, we expect that the values of CSR and PRI are big and that the VI value is small.