Skip to main content

Table 4 10-fold cross validation performance among the best performing model after C and \( \boldsymbol{\gamma} \) optimization and the model using default C and \( \boldsymbol{\gamma} \). (F: F-1 score; P: precision; R: recall; for the categories that do not indicate the metric, F-1 score are used)

From: Optimization on machine learning based approaches for sentiment analysis on HPV vaccines related tweets

Model

Model using default C and \( \boldsymbol{\gamma} \)

Best model using optimized \( \boldsymbol{\gamma} \) only

Best model using optimized C and \( \boldsymbol{\gamma} \)

Micro-averaging

F

0.73

0.7352

0.7442

Macro-averaging

P

0.5477

0.6889

0.6873

 

R

0.4576

0.5095

0.5142

F

0.4986

0.5858

0.5883

Unrelated

0.8044

0.8538

0.8633

Neutral

0.5792

0.6330

0.6470

Positive

0.6528

0.7239

0.7255

NegSafety

0.7006

0.7641

0.7617

NegEfficacy

0

0.4138

0.4068

NegCost

0

0.5

0.5

NegResistant

0

0

0

NegOthers

0.155

0.5144

0.5403