Feature selection improves the classification performance of machine learning models. It also identifies the important features and eliminates those with little significance. Furthermore, feature selection reduces the dimensionality of training and testing data points. This study proposes a feature selection method that uses a multivariate sample similarity measure. The method selects features with significant contributions using a machine-learning model. The multivariate sample similarity measure is evaluated using the University of California, Irvine heart disease dataset and compared with existing feature selection methods. The multivariate sample similarity measure is evaluated with metrics such as minimum subset selected, accuracy, F1-score, and area under the curve (AUC). The results show that the proposed method is able to diagnose chest pain, thallium scan, and major vessels scanned using X-rays with a high capability to distinguish between healthy and heart disease patients with a 99.6% accuracy.
Improper management of reactive power in a power network could lead to voltage instability. This paper presents a well-detailed study on voltage instability due to violation of power equilibrium in a power network and introduces a new voltage stability pointer (NVSP). The proposed NVSP is developed from a reduced 2-bus interconnected network to predict the sensistivity of voltage stability to reactive power variation. The simulation results from MATLAB were evaluated on IEEE 14-bus test system. The contingency ranking was achieved by varying the reactive power on the load buses to its maximum loading limit. The maximum reactive power point was taken at each load bus and the critical lines were ranked according to their vulnerability to voltage collapse. The results were compared with other notable voltage stability indices. The results prove that the NVSP is an essential tool in predicting voltage collapse.
This paper presents an assessment of the levels of total harmonic distortion (THD) in buck-boost DC-AC converters using triangular wave and saw-tooth unipolar based-modulation schemes. This paper seeks to identify a better technique for mitigating the total harmonic distortion on buck-boost DC-AC converters under unipolar carrier-based modulation schemes. This was achieved by subjecting the buck-boost DC-AC converter under triangular wave-based and saw-tooth based-unipolar modulation schemes. The voltage and current output of the buck- boost DC-AC converter under each scheme was analysed using a power GUI Fast Fourier Transform (FFT) analytical tool resident in the MATLAB Simulink environment unlike with the conventional scheme of computing the percentage of THD. The test system was obtained by a combination of DC-DC buck-boost converter, H-bridge based-insulated unipolar gate transistors, and a logic control unit. It was realized that THD of 0.2865%, peak output voltage of 294.1V and current of 9.805A were obtained by using the saw-tooth based-unipolar modulation scheme, whereas a THD of 0.1479%, peak output voltage of 297.4V and current of 9.53A were obtained by using the triangular wave based-bipolar modulation scheme on the same Buck-boost DC-AC converter circuit. The results imply a high power factor utilization and low power loss in the triangular wave based-unipolar modulation scheme compared to the saw-tooth based-unipolar modulation technique for improving the performance characteristics of the buck-boost converter system. This study showed that power drives and heavy load machines based-power electrical loads are required to use the saw-tooth based-unipolar modulation (STBUM) scheme for high current and low THD%, whereas sensitive power electrical loads such as hospital equipment and communication industries based-power electronic devices are required to use the triangular wave-based unipolar modulation (TWBUM) scheme due to low current and THD%.
Predictive modeling of asthma characterized by the systematic use of Machine Learning and Deep Learning techniques to develop classification/prediction models is a vital tool which is being deployed in most of the computer mediated decision making processes. Spirometry, being one of the most commonly used lung function tests, helps in the diagnosis and continuous monitoring of asthma and is recommended by both the national and international guidelines for the management of the disease when compared to other pulmonary function tests. It has been found to be more reliable because it has more parametric values. Despite the generalization of the respiratory equations in spirometry with respect of selected ethnic groups, the equation yields a considerable difference when compared to the spirometric readings in the general population. In an effort to overcome such differences that deviate from actual observations, in this paper, we have proposed a neural network model that can output a vector of Tiffeneau-Pinelli Index. The neural network model for the prediction of Tiffeneau-Pinelli index was able to reproduce a vector of indices that very closely approximated the actual observed values with a very low estimated error with an optimized radial basis fit neural net. This can be used as a reliable means to estimate some of the vital lung function parameters irrespective of the differences in the general population.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.