Semisupervised Learning Based Disease-Symptom and Symptom-Therapeutic Substance Relation Extraction from Biomedical Literature

Table 2

The initial results on the disease-symptom test set. Method 1 integrates three classifiers (the feature kernel, graph kernel, and tree kernel) with the same weight while Method 2 integrates them with a weight ratio of 4 : 4 : 2.

Method			-score	AUC
Feature kernel	91.38	62.11	73.95	87.13
Graph kernel	93.87	59.77	73.04	87.21
Tree kernel	69.10	62.89	65.85	73.37
Method 1	92.05	63.28	75.00	89.47
Method 2	92.81	60.55	73.29	89.74