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

Table 11

LSTM classification performance using embeddings over the BioText Berkeley dataset.

Embeddings	Accuracy
	,		,		,		,
	Training	Validation	Training	Validation	Training	Validation	Training	Validation
GloVe_C	0.8518	0.8563	0.8474	0.8035	0.8698	0.8822	0.8663	0.8410
GloVe_W	0.8581	0.8492	0.8396	0.7977	0.8682	0.8651	0.8582	0.8006
GloVe_Merged	0.8230	0.8246	0.8461	0.7801	0.8376	0.8182	0.8412	0.8208
CE	0.8880	0.8628	0.8513	0.8271	0.8737	0.8765	0.8787	0.8430
WE	0.8758	0.8552	0.8474	0.8183	0.8837	0.8306	0.8628	0.8377
Merged	0.8321	0.8283	0.8486	0.8039	0.8482	0.8294	0.8436	0.8259

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