where α i is the positive Lagrange multiplier, x i is the support vector (a total of N) and K(x i, x) is the kernel decision function.
Subsequently, 9 distinct types of predictive models are fitted using the reduced data set: logistic regression (LR), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machines (SVMs; using linear, radial basis function and polynomial kernels), decision tree (DT), random forest (RF), and stochastic gradient boosting (SGB).
The reduced data set of 24 parameters is fitted using 9 distinct types of predictive models: logistic regression, linear discriminant analysis, quadratic discriminant analysis, support vector machines (using linear, radial basis function and polynomial kernels), decision tree, random forest and stochastic gradient boosting.
The decision-making kernel of the decision support system is implemented as a multi-agent-system (MAS).
A number of machine learning approaches can be used to perform this comparison: statistical approaches such as hidden Markov models, neural networks, kernel methods, and decision trees.
Due to its superior performance on prediction accuracy, we chose RBF-SVM (support vector machine with the radial basis function kernel as the decision function) to identify the mature miRNA location.
The model is used as the kernel of the cognitive decision engine (CDE).
The presented algorithm is outlined in a hierarchical way and embedded into dedicated agents as decision-making kernel.
Therefore, an RBF kernel (a non-linear decision plane) is used as it yields a greater accuracy than a linear kernel, as illustrated in Figs. 5 and 6.
Results are benchmarked against more traditional methods under consideration for commercial applications including linear discriminant analysis, logistic regression, k nearest neighbor, kernel density estimation, and decision trees.
