Similarity-Based Method with Multiple-Feature Sampling for Predicting Drug Side Effects

Figure 1

Procedures of multiple-feature sampling scheme to extract essential features from a drug–side effect pair. For a pair of drug and side effect , drugs having the side effect are extracted from the training dataset. The association scores between and these drugs constitute a candidate feature list. The discrete strategy selects discrete values in such list as features, and the continuous strategy picks up some top values in this list as features.