Skip to main content
Fig. 5 | BMC Medical Informatics and Decision Making

Fig. 5

From: Disease causality extraction based on lexical semantics and document-clause frequency from biomedical literature

Fig. 5

Experimental Results: a For 125 diseases, \( \alpha DCFC \) extracted 2.7 times more causal relationship compared with dRiskKB, and covered 79% of the existing causal relationships. b When the ranks of the causality strength are compared, the proposed method has a higher rank correlation coefficient with the document frequency than that of dRiskKB; the document frequency represents the causal relationships. c Trend lines of the causality strength of two methods show that \( \alpha DCFC \) is more similar to the document frequency than dRiskKB

Back to article page