Table 12 Detailed cross-validation results for STRATEGY 1 and the ANN-LSA model (Stackoverflow dataset).

From: Analysis of the retraining strategies for multi-label text message classification in call/contact center systems

k

Training data

Testing data

Accuracy

Emotica

Accuracy

Emotica

[%]

[%]

[%]

[%]

1

99.44

97.03

96.14

86.25

2

99.42

96.76

95.00

82.16

3

99.48

97.44

95.00

82.97

Average

99.45

97.08

95.38

83.79

SD

0.02

0.28

0.54

1.77

  1. Significant values are in bold.
Back to article page