|
| fp | Forced prediction. Forcing the algorithm to make a prediction at each intermediate point, ignoring nonpredictable points |
| cm | Close motifs. Analyzing the number of “close motifs” of each intermediate point |
| en | Entropy. Analyzing the entropy of the sets of possible predicted values of each intermediate point |
| ap | A priori. Using a priori information |
| lr | Logistic regression. Using a logistic regression classifier |
| svm | Support vector machine. Using an SVM classifier employing polynomial kernel functions |
| dt | Decision tree. Applying a decision tree employing an entropy criterion without any restriction on the tree depth to the features of the sets of possible predicted values |
| knn | K-nearest-neighbors. Applying a k-nearest-neighbors classifier is applied to the features of the sets of possible predicted values; k-nearest-neighbors value is equal to 3 |
| mlp | Multilayer perceptron. Applying a multilayer perceptron containing 8 hidden layers and utilizes a sigmoidal activation function to the features of the sets of possible predicted values |
| adb | AdaBoost. Assembling logistic regression classifiers using AdaBoost |
| adblr | AdaBoost + logistic regression. Logistic regression classifiers are assembled with AdaBoost |
| adbsvm | AdaBoost + SVM. Assembling SVM classifiers using AdaBoost |
| lrs | Logistic regression, stacking. Stacking of logistic regression classifiers |
| lrv | Logistic regression, voting. Assembling logistic regression classifiers by voting |
| dbscan | DBSCAN. Clustering the sets of possible predicted values at each intermediate point using DBSCAN and calculating the growth rate of the total number of clusters over several consecutive points |
| wshrt | Wishart. Clustering the sets of possible predicted values at each intermediate point using Wishart clustering and calculating the growth rate of the total number of clusters over several consecutive points |
| cwat | Compare weighted average, trajectories. Comparing the main predicted trajectory with a weighted average of other predicted trajectories |
| amd | Average modulus of difference. Calculating an average of modulus of a difference |
| wamd | Weighted average modulus of difference. Calculating a weighted average of modulus of a difference |
| wamdt | Weighted average modulus of difference, trajectory. Comparing a weighted average of modulus of a difference of the main trajectory to that of other trajectories |
| rg | Rapid growth. The spread grows monotonically over several consecutive points |
| rgdbscan | Rapid growth, DBSCAN. DBSCAN is used to obtain centers of clusters of possible predicted values |
| rgwshrt | Rapid growth, Wishart clustering. Wishart clustering is used to obtain centers of clusters of possible predicted values |
|
