Skip to main content

Table 7 Results for Swedish

From: Recent advances in Swedish and Spanish medical entity recognition in clinical texts using deep neural approaches

SourceFeatureAlgorithmPrecisionRecallF1-score
EHRWGlove76.0374.4275.22
  word2vec75.9175.4475.68
  FastText76.3574.9075.62
 LGlove76.0472.1074.02
  word2vec74.4474.4974.46
  FastText75.2572.9974.10
GenMedWGlove73.1471.1572.13
  word2vec74.8368.5571.56
  FastText74.0969.6571.80
 LGlove74.7667.6771.03
  word2vec74.0568.8971.38
  FastText75.4869.5172.37
GenWGlove73.7967.6070.56
  word2vec72.5070.3371.40
  FastText76.1868.4972.13
 LGlove73.1570.0571.57
  word2vec72.4768.7670.56
  FastText73.9668.0170.86
EHRW,Lword2vec74.6477.4976.04
EHRW+Lword2vec74.4376.2675.34
EHRW-Lword2vec73.4550.9560.17
  1. Embeddings of single base-units on top and, below, with combined base-units