Skip to main content

Table 2 – Performance of our proposed methods on the 2010 i2b2/VA subset (5-fold cross validation)

From: Integrating shortest dependency path and sentence sequence into a deep learning framework for relation extraction in clinical text

Features

Precision (%)

Recall (%)

F-measure (%)

∆ (%)

Sentence Sequence only

74.01

69.79

71.84

–

+SDP (Word Sequence)

74.20

72.84

73.51

1.67

+SDP (Word Sequence + Relation Type)

75.69

73.03

74.34

2.50