Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: A meta-learning approach for B-cell conformational epitope prediction

Figure 4

Cascade generalization architecture. The conformational epitope predictors and linear epitope predictors all served at Level 0 as the base predictors. We placed k-NN, C4.5, ANN, and SVM sequentially from Levels 1 to 4 as meta learners. Each meta learner generalized the output from the previous level to meta knowledge in the form of meta features. The meta features and base features propagated sequentially to the successive level as input to the subsequent meta learner. The top-level meta learner, SVM, produced the final meta classification.

Back to article page