Skip to main content

Table 5 Results of statistical analysis (p-values)

From: Identifying tweets of personal health experience through word embedding and LSTM neural network

Classifier Accuracy Precision
(PET)
Recall
(PET)
F1
(PET)
ROC/AUC
Logistic Regression 2.52 × 10− 08 1.85 × 10−09 5.48 × 10− 09 5.87 × 10−10 1.46 × 10− 09
Decision Tree 1.80 × 10− 04 1.51 × 10− 04 6.99 × 10− 06 1.92 × 10− 06 1.16 × 10− 05
KNN 8.08 × 10− 05 6.22 × 10− 05 1.40 × 10 − 03 8.50 × 10− 05 1.29 × 10 − 04
SVM 1.17 × 10− 08 4.61 × 10− 08 1.74 × 10− 04 7.89 × 10− 07 5.02 × 10− 06
BoW + Logistic Regr. 4.26 × 10 − 04 2.22 × 10 − 04 1.79 × 10− 04 9.85 × 10 − 05 2.12 × 10− 05
  1. The highest values are in boldface