Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Genome-wide sequence-based prediction of peripheral proteins using a novel semi-supervised learning technique

Figure 2

Two step strategy used in Positive-Unlabeled learning. In the first step, a set of reliable negative examples are identified using a spy technique. During the spy technique, some positive examples are included in the U set as spies. After the classification, a threshold is chosen such that all the spies are classified as positive and the ones below the threshold form the reliable negative (RN) set. In the second step, the RN set and the P set are used to build a classifier.

Back to article page