Table 5 Results of the KM-DBSCAN data reduction compared to other methods from the related work on the banana dataset.

From: KM-DBSCAN: an enhanced density and centroid based border detection framework for data reduction towards green AI

Method

Accuracy

Loss/Gain

in Accuracy

Reduction

Ratio

SVs

Reduction ratio

Speedup

Whole dataset

0.9044

–

–

–

–

FIFDR9

0.8770

− 0.0274

0.684

0.315

4.990

CBCH10

0.8990

− 0.0054

0.706

1.024

1.541

BPLSH11

0.8818

− 0.0226

0.247

0.858

5.139

DBI14

0.9000

− 0.0044

0.500

1.028

1.980

BRI14

0.8930

− 0.0114

0.200

0.470

9.360

BRIX14

0.8800

− 0.0244

0.100

0.068

51.620

SVO14

0.8980

− 0.0064

0.221

0.996

5.230

SVOX14

0.9020

− 0.0024

0.133

0.260

17.140

Proposed method

0.91074

+ 0.00634

0.0372

0.05436

98.2