Research Article
Biomedical Text Classification Using Augmented Word Representation Based on Distributional and Relational Contexts
Table 3
Word similarity performance with
and
.
| Word embeddings | BioSimLex | BioSimVerb | MEN | MTurk | RG65 | RW | SCWS | SimLex999 | TR9856 | UMNSRS-rel | UMNSRS-sim | VERB143 | WS353 | WS353R | WS353S |
| GloVe_W | 0.39595 | 0.02102 | 0.08633 | 0.18420 | −0.10868 | 0.19159 | 0.33133 | 0.09255 | 0.14876 | 0.02863 | 0.11977 | 0.20532 | 0.27138 | 0.1856 | 0.28586 | GloVe_C | 0.38942 | 0.02463 | 0.08613 | 0.18368 | −0.11569 | 0.19241 | 0.33356 | 0.09008 | 0.14976 | 0.02911 | 0.11795 | 0.20424 | 0.27077 | 0.18893 | 0.29353 | GloVe_Merged | 0.39252 | 0.0167 | 0.08978 | 0.18971 | −0.1144 | 0.19416 | 0.32867 | 0.08579 | 0.14863 | 0.03666 | 0.13204 | 0.16467 | 0.26551 | 0.18294 | 0.29706 | WE | 0.45936 | 0.13931 | 0.22985 | 0.34154 | 0.16416 | 0.17521 | 0.37285 | 0.17074 | 0.16225 | 0.21659 | 0.31450 | 0.23520 | 0.23407 | 0.18545 | 0.31335 | CE | 0.45911 | 0.14452 | 0.22756 | 0.34159 | 0.15423 | 0.16408 | 0.37965 | 0.16941 | 0.16867 | 0.23040 | 0.31535 | 0.24051 | 0.26612 | 0.22927 | 0.33163 | Merged | 0.46832 | 0.15000 | 0.23151 | 0.33007 | 0.12302 | 0.15321 | 0.37304 | 0.15882 | 0.16605 | 0.22527 | 0.30895 | 0.208215 | 0.22719 | 0.19314 | 0.29428 |
|
|
Bold means the best performance in the case of each dataset.
|