Skip to main content

Table 3 Comparison of trigger detection performance with different word embedding

From: Bidirectional long short-term memory with CRF for detecting biomedical event trigger in FastText semantic space

Semantic Space

F-score(%)

Precision(%)

Recall(%)

Random Embedding

73.72

76.36

71.25

Glove

74.84

79.70

70.54

Doc2Vec

76.03

78.78

73.47

Word2Vec

76.71

79.61

74.02

Ours(FastText)

78.08

77.89

78.28