Research Article

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

Table 2

Word similarity performance with and .

Word embeddingsBioSimLexBioSimVerbMENMTurkRG65RWSCWSSimLex999TR9856UMNSRS-relUMNSRS-simVERB143WS353WS353RWS353S

GloVe_W0.370860.024250.088330.19136−0.122190.189130.327340.089830.147760.020510.120930.203390.265370.174540.28847
GloVe_C0.377040.02340.088820.19213−0.114860.189790.324710.089230.148500.022980.122160.202140.264710.178470.29069
GloVe_Merged0.380460.016840.089010.19784−0.110800.191420.322500.085930.146830.023330.128010.174830.263120.176160.29588
WE0.386560.091450.160120.193780.086060.166190.361040.157730.161010.234480.333790.177400.281740.248850.30282
CE0.38640.089320.157780.196370.068700.154680.362620.166560.164860.250850.340710.167460.283440.239620.31748
Merged0.419830.103330.178760.225600.133490.167650.370260.164200.159740.240830.343550.196620.273400.242280.29602

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