Differential evolution trained kernel principal component WNN and kernel binary quantile regression: Application to banking

Interval combined prediction of mine tunnel's air volume considering multiple influencing factors

Continuous monitoring and accurate measurement of required air volume in mine tunnels are crucial phenomena for mine safety However, air volume fluctuates and can become unstable which can lead to biased measurement in underground environment. In this paper, to accurately measure the mine tunnel air volume, the tunnel air volume, and related ventilation parameters are consistently monitored, and the real monitoring data is converted to interval numbers for representation. These interval numbers are then preprocessed using an Interval-type Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(In-CEEMDAN) to extract the essential features of the data. Then, the monitored data is processed using the phase space reconstruction technique to identify the most relevant influencing factors related to the air volume. The tunnel air volume and influencing factors are then input into different neural networks for air volume prediction. To further improve prediction accuracy, the predicted values of wind volume intervals from the single prediction method are transformed into triangular fuzzy numbers, and the generalized induced ordered weighted average operator is introduced for the combination of prediction results. The grey correlation method is selected as the optimization criterion, and the preference coefficients are used to transform the multi-objective optimization problem into a single-objective optimization problem. In order to reduce the prediction error, the L2 paradigm is combined with the gray correlation to construct a complete interval combination type air volume prediction model which considers multiple influencing factors. Finally, a sensitivity analysis was carried out to analyze the values of the preference coefficients in the model, and the final range of values was given. Experimental analysis using data from a coal mine in Inner Mongolia showed that the method could reduce Combined Weighted Mean Absolute Error(CWMAE) to a maximum of 5.0384, Combined Weighted Root of Mean Squares Error(CWRMSE) to 6.8889, and Combined Weighted Mean Absolute Percentage Error(CWMAPE) to 1.4756, which indicates that the method proposed in this study can effectively improve the prediction accuracy of the mine tunnel air volume.

Evolutionary sequential genetic search technique-based cancer classification using fuzzy rough nearest neighbour classifier

Cancer is one of the deadly diseases of human life. The patient may likely to survive if the disease is diagnosed in its early stages. In this Letter, the authors propose a genetic search fuzzy rough (GSFR) feature selection algorithm, which is hybridised using the evolutionary sequential genetic search technique and fuzzy rough set to select features. The genetic operator's selection, crossover and mutation are applied to generate the subset of features from dataset. The generated subset is subjected to the evaluation with the modified dependency function of the fuzzy rough set using positive and boundary regions, which act as a fitness function. The generation and evaluation of the subset of features continue until the best subset is arrived at to develop the classification model. Selected features are applied to the different classifiers, from the classifiers fuzzy-rough nearest neighbour (FRNN) classifier, which outperforms in terms of classification accuracy and computation time. Hence, the FRNN is applied for performance analysis of existing feature selection algorithms against the proposed GSFR feature selection algorithm. The result generated from the proposed GSFR feature selection algorithm proved to be precise when compared to other feature selection algorithms.

Kernel association for classification and prediction: a survey

Kernel association (KA) in statistical pattern recognition used for classification and prediction have recently emerged in a machine learning and signal processing context. This survey outlines the latest trends and innovations of a kernel framework for big data analysis. KA topics include offline learning, distributed database, online learning, and its prediction. The structural presentation and the comprehensive list of references are geared to provide a useful overview of this evolving field for both specialists and relevant scholars.