Skip to main content

Table 10 Token level errors in percentages by type (correctly and incorrectly tagged) and possible error sources

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

 SwedishSpanish
 OOEVAAVOOTVOOEVAAVOOTV
Correct8.049.5138.384.099.1339.37
Incorrect11.3010.4366.384.5613.6958.94
Both6.9033.5040.912.3440.1133.45