Skip to main content

Table 9 F1 scores (the highest scores italicized)

From: Developing an online hate classifier for multiple social media platforms

 

Simple features

BOW

TF-IDF

Word2Vec

BERT

All featuresa

LR

0.062

0.764

0.768

0.828

0.891

0.892

NB

0.130

0.505

0.606

0.601

0.885

0.868

SVM

0.066

0.487

0.648

0.765

0.892

0.883

XGBoost

0.400

0.765

0.774

0.880

0.916

0.924**

FFNN

0.064

0.770

0.769

0.847

0.893

0.894

KBC

n/a

n/a

n/a

n/a

n/a

0.388

BOC

n/a

n/a

n/a

n/a

n/a

0.084

  1. ** Significant at p < 0.001 (McNemar’s test comparing predictions from XGBoost-BERT and XGBoost-All
  2. aThe features are concatenated into one big vector for each instance and used as input to the classifiers
Back to article page