Table 3 Performance of MRAM in term of likelihood lift with different subsets of data features and the three sets of negative examples with different sizes n1, n2, n3.

From: Novel drug target identification for the treatment of dementia using multi-relational association mining

Experiment

n 1  = 221

n 2  = 500

n 3  = 1,000

Exp1: All data features excluding the topological data features

0.391

0.412

0.211

Exp3: All data features excluding the GO data feature

0.398

0.443

0.220

Exp4: All data features excluding the Reactome data feature

0.408

0.436

0.215

Exp5: All data features excluding the InterPro data feature

0.411

0.449

0.237

Exp6: All of investigated data feature

0.433

0.469

0.256

  1. The best results obtained are labeled in bold.
Back to article page