Biomedical Text Classification Using Augmented Word Representation Based on Distributional and Relational Contexts

Table 16

BiLSTM classification performance using embeddings over the PubMed_20k_RCT dataset.

Embeddings	Accuracy
	,		,		,		,
	Training	Validation	Training	Validation	Training	Validation	Training	Validation
GloVe_C	0.7588	0.7478	0.7705	0.7650	0.7593	0.7573	0.7706	0.7631
GloVe_W	0.7569	0.7464	0.7707	0.7699	0.7601	0.7470	0.7719	0.7680
GloVe_Merged	0.7306	0.7279	0.7411	0.7381	0.7328	0.7283	0.7414	0.7410
CE	0.7545	0.7362	0.7850	0.7667	0.7598	0.7561	0.7909	0.7687
WE	0.7576	0.7481	0.7898	0.7685	0.7667	0.7486	0.7941	0.7727
Merged	0.7326	0.7285	0.7501	0.7449	0.7396	0.7315	0.7786	0.7532

Bold means the best performance in the case of each dataset.