Search

Gestures are one of the most important agents for human-computer interaction. They play a mediator role between human intention and the control of machines. Electromyography (EMG) pattern recognition has been studied for gesture recognition for years to control of prostheses and rehabilitation systems. EMG data gives information about the electrical activity related to muscles. It is obtained from the arm and helps to understand hand gestures. For this work, hand gesture data taken from UCI2019 EMG dataset obtained from myo Thalmic armband were classified with six different machine learning algorithms. Artificial Neural Network (ANN), Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), Naive Bayes (NB), Decision Tree (DT) and Random Forest (RF) methods were applied for comparison based on several performance metrics which are accuracy, precision, sensitivity, specificity, classification error, kappa, root mean squared error (RMSE) and correlation. The data belongs to seven hand gestures. 700 samples from 7 classes (100 samples per group) were used in the experiments. Splitting ratio in the classification was 0.8-0.2, i.e. 80% of the samples were used in training and 20% of data were used in the testing phase of the classifier. NB was found to be the best among other methods because of high accuracy. Classification accuracy varies between 97.52% to 100% for each gesture. Considering the results of the performance metrics, it can be said that this study recognizes and classifies seven hand gestures successfully in comparison with the literature. The proposed method can easily be used for human-machine interaction and smart device controlling like prosthesis, wheelchair, and smart entertainment applications. WOS:000704721300002

Today, the world's primary energy demand has been met by the burning of fossil-based fuels at a rate of 85%. This dominant use of fossil-based fuels has led to an accelerating increase in the release of greenhouse gases (GHG) all across the world. The largest share in total GHG emissions belongs to the electricity and heat production sector with a rate of 25%. With this viewpoint, this paper is aiming to forecast the GHG emissions (CO2, CH4, N2O, F-gases, and total GHG) using deep learning (DL), support vector machine (SVM), and artificial neural network (ANN) algorithms from the electricity production sector in Turkey. The dataset is supplied from the Turkish Statistical Institute and covers the years 1990-2018. In the study, the last four years (2015-2018) is being forecasted. To evaluate the performance success of the algorithms, five metrics (RMSE, MBE, rRMSE, R-2, and MAPE) are discussed in detail. In the results, this research is reporting that all algorithms used in the study are giving separately satisfying results for the forecasting of GHG emissions in Turkey. Based on the forecasting results, it is seen that the highest R-2 value for the emissions varies from 0.861 to 0.998 and all results are categorized as excellent in terms of rRMSE (all rRMSE values < 10%). Besides, MBE changes between -2.427 and 2.235, and all MAPE values are smaller than 1.2%. Total GHG emission is forecasted in DL algorithm with very satisfied R-2, RMSE, MBE, rRMSE, and MAPE of 0.998, 2.046, 0.419, 0.406%, and 0.021%, respectively. On the other hand, CO2 accounted for 69.05% of total GHG emissions of Turkey in 1990 but rising by 80.48% in the year 2018. In comparison with those of 1990, electricity production and total GHG emissions of Turkey in 2018 increased by 429.7% and 137.4%, respectively. Total GHG emission corresponding to electricity production is calculated to be 0.3813 Mt-total GHG/MWh in 1990 and 0.1709 Mt-total GHG/MWh in 2018. In conclusion, GHG emissions have recently increased at a high rate, but it is noticed that this increase is considerably higher as compared to the increase in energy production for Turkey. (C) 2020 Elsevier Ltd. All rights reserved. WOS:000611934600022 2-s2.0-85097710385