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

Table 7

Concept categorization performance with and .

Word embeddings	AP	BLESS	Battig	ESSLI_1a	ESSLI_2b	ESSLI_2c	Ohta-10-bio-words
GloVe_W	0.22636	0.22	0.11393	0.43182	0.525	0.4	0.48275
GloVe_C	0.22388	0.23	0.11049	0.40909	0.5	0.35556	0.45689
GloVe_Merged	0.23880	0.25	0.11948	0.47727	0.525	0.4	0.44827
WE	0.26119	0.33	0.12388	0.65909	0.525	0.35556	0.49138
CE	0.25373	0.295	0.12235	0.56818	0.6	0.37778	0.48276
Merged	0.26119	0.305	0.12483	0.5	0.525	0.4	0.49138

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