Search

Using Hybrid optimization algorithms for nonlinear systems analysis is a novel approach. It is a powerful technique that uses the exploitation ability of one algorithm and the exploration ability of another algorithm, to find the best solution. Literature survey reveals that hybrid algorithms not only show quality response but also give faster convergence of error for nonlinear systems. In this paper, hybrid optimization techniques based proportional integral derivative (PID) controller is used for benchmark problems: Continuous stirred tank reactor (CSTR), Inverted pendulum and blood glucose system. Two recent hybrid algorithms: Particle swarm optimization‐Gravitational search algorithm (PSO‐GSA) and Particle swarm optimization‐Grey wolf algorithm (PSO‐GWO) are implemented to control the temperature and concentration of CSTR, pendulum angle of inverted pendulum, glucose concentration and insulin level of blood glucose system. In PID and PSOGWO algorithms, the exploration abilities of GSA and GWO combined with the exploitation ability of PSO have been used. The performance of these algorithms is then compared with individual PSO, GSA, and GWO algorithms proving their superiority. Stability is ensured using the Lyapunov approach while the robustness of the systems is checked using the parameter perturbation technique. Simulation results show substantial improvement in the performance of these systems by using these meta‐heuristic hybrid optimization techniques. A comparative analysis of these algorithms has also been done. [ABSTRACT FROM AUTHOR]

In today's world multiple type of datasets are available, they are mainly divided into two type namely classification and regression; based on the dataset type machine learning models are applied. There is a technique called bagging in which multiple models are trained simultaneously and produce results on testing dataset. These results are then clubbed to vote in the majority of the same outputs produced. This output is then considered as final output. The bagging technique used in this study is Random Forest (RF) algorithm, where the multiple models are replaced with decision trees. The RF algorithm requires multiple parameters to produce optimum output. These parameters are n_estimators, max_ samples, max_features etc. Though RF algorithm produces favourable results with default values of parameters, but to improve the efficiency of RF algorithm tuning of parameters is preferred. In this study, parameter n_estimators is tuned, on basis of the length of dataset to produce improved results as compared to the default parameters value. Further the proposed tuning method has been applied to RF algorithm and improvement of class prediction efficiency on different datasets is measured in terms of accuracy, precision, recall, F1-score. Also, implementation of ROC curves and auc is performed to depict the improvement of RF algorithm for datasets under consideration, for instance increase in auc values from 0.915 to 0.932 for Spine (2 Classes) dataset and 0.705 to 0.720 for Haberman's survival dataset. [ABSTRACT FROM AUTHOR]