Skip to main content

Table 4 Results of the best performing dimensionality reduction techniques based on F-score

From: Spike pattern recognition by supervised classification in low dimensional embedding space

Method

Sensitivity

FP/min

F-score

LLTSA

0.96

0.10

0.77

LPP

0.97

0.11

0.76

NPE

0.97

0.12

0.74

PCA

0.96

0.13

0.72

MCML

0.96

0.13

0.72

SPE

0.97

0.14

0.71

DiffusionMaps

0.97

0.15

0.70

LLE

0.86

0.14

0.66

MVU

0.86

0.14

0.65

NCA

0.98

0.20

0.64

LandmarkIsomap

0.93

0.19

0.64

FactorAnalysis

0.94

0.23

0.60

Isomap

0.95

0.23

0.60