Advances in Fuzzy Systems

Research Article

Fuzzy Lattice Reasoning for Pattern Classification Using a New Positive Valuation Function

Table 3

Recognition results along with relative ranking by different methods over 6 benchmarks.
Data set∖AlgorithmSOMFuzzy-ARTGRNN F L R 𝑥 F L R 𝑎 F L R 𝑙
Columbia Image 4 3 . 8 7 ( 4 ) 4 0 . 8 1 ( 5 ) 9 0 . 1 6 ( 3 ) 9 4 . 0 3 ( 2 ) 𝟗 𝟒 . 𝟑 𝟔 ( 𝟏 ) 𝟗 𝟒 . 𝟑 𝟔 ( 𝟏 )
Image Segmentation 7 9 . 8 6 ( 5 ) 8 8 . 7 1 ( 4 ) 9 2 . 4 3 ( 3 ) 𝟗 𝟑 . 𝟎 𝟎 ( 𝟏 ) 9 2 . 7 1 ( 2 ) 𝟗 𝟑 . 𝟎 𝟎 ( 𝟏 )
Pen-based recognition 7 9 . 2 4 ( 6 ) 6 5 . 7 0 ( 5 ) 8 1 . 8 5 ( 2 ) 8 1 . 4 8 ( 3 ) 8 0 . 8 5 ( 4 ) 𝟖 𝟏 . 𝟗 𝟎 ( 𝟏 )
Letter recognition 6 8 . 6 5 ( 5 ) 6 5 . 2 2 ( 6 ) 8 2 . 3 3 ( 4 ) 8 4 . 2 6 ( 2 ) 8 3 . 4 2 ( 3 ) 𝟖 𝟒 . 𝟒 𝟖 ( 𝟏 )
Semion hand recognition 7 1 . 5 6 ( 4 ) 6 8 . 2 7 ( 5 ) 7 8 . 3 4 ( 3 ) 7 8 . 9 7 ( 2 ) 7 8 . 9 7 ( 2 ) 𝟕 𝟗 . 𝟏 𝟖 ( 𝟏 )
Optical recognition of handwritten digits 8 7 . 8 7 ( 5 ) 8 8 . 7 6 ( 4 ) 𝟗 𝟒 . 𝟒 𝟗 ( 𝟏 ) 9 3 . 2 7 ( 3 ) 9 3 . 6 0 ( 2 ) 9 3 . 2 7 ( 3 )