Figure 1 | Scientific Reports

Figure 1

From: Adaboost-SVM-based probability algorithm for the prediction of all mature miRNA sites based on structured-sequence features

Figure 1

The schematic of the overall method. MiRFinder consists of five steps, namely, dataset construction, construction of structured-sequence, biological-related features extraction, feature selection, class imbalance problem and classifier construction. For features selection, three algorithms including information gain, chi-square and relief were investigated. For class imbalance problem, the AdaBoost algorithm was adopted. Moreover, for classifier construction, the AdaBoost and SVM were used. SVM is used as the weak classifier, and AdaBoost is used as the strong classifier.

Back to article page