Skip to main content

Table 3 CK-36-SEQ

From: Compression-based classification of biological sequences and structures via the Universal Similarity Metric: experimental assessment

CK-36-SEQ UCD NCD CD
  UPGMA NJ UPGMA NJ UPGMA NJ
Gzip 0.8500 0.9030 0.8500 0.7600 0.9030 0.7600
Bzip2 0.8585 0.6970 0.8265 0.7236 0.9030 0.7462
PPMd16 0.9030 0.7827 0.9030 0.7600 0.9030 0.7827
PPMd8 0.9030 0.8407 0.9030 0.7600 0.5501 0.5456
PPMd4 0.9030 0.7069 0.9030 0.7600 0.5389 0.5517
PPMd2 0.8500 0.7069 0.8500 0.7069 0.5449 0.5386
Huffman 0.8188 0.7609 0.8645 0.7609 0.8161 0.7066
Ac fast 0.8392 0.6770 0.8392 0.6770 0.8410 0.7324
Rc fast 0.8706 0.7275 0.8706 0.7275 0.8734 0.7268
Ac med. 0.8518 0.6718 0.8518 0.6718 0.8584 0.7288
Rc med. 0.8780 0.7674 0.8466 0.6825 0.7936 0.6621
Ac slow 0.8518 0.7009 0.8392 0.7009 0.8584 0.7268
Rc slow 0.9030 0.7357 0.9030 0.7188 0.8734 0.7288
BwtRleHuff 0.8319 0.6970 0.8585 0.8442 0.7840 0.7440
BwtMtfRleHuff 0.8501 0.8355 0.8253 0.6770 0.7971 0.6692
BwtRleAc fast 0.8706 0.7008 0.8706 0.8026 0.9030 0.7515
BwtMtfRleAc fast 0.8382 0.7217 0.8126 0.7336 0.8706 0.8001
BwtRleRc fast 0.8706 0.7343 0.8706 0.6889 0.9030 0.7950
BwtMtfRleRc fast 0.8500 0.7217 0.8500 0.7217 0.8500 0.7515
BwtRleRc med. 0.8500 0.7074 0.8706 0.7336 0.9066 0.8762
BwtMtfRleRc med. 0.8252 0.6951 0.8252 0.6951 0.8252 0.6657
BwtRleRc slow 0.9030 0.7357 0.8585 0.7297 0.9030 0.7084
BwtMtfRleRc slow 0.8706 0.7480 0.8126 0.7910 0.8706 0.6825
BwtWavelet 0.8706 0.6887 0.8706 0.7993 0.8647 0.7066
  1. Experimental results for the CK-36-SEQ data set, with the UCD (left), NCD (middle), and CD (right) distance. For each compression algorithm, we report the F-measure for both UPGMA and NJ methods.