ABSTRACT
Background: Computational intelligence (CI) based prediction models increase the efficient and effective utilization of resources for wind prediction. However, the traditional recurrent neural networks (RNN) are difficult to train on data having long-term temporal dependencies, thus susceptible to an inherent problem of vanishing gradient. This work proposed a method based on an advanced version of RNN known as long short-term memory (LSTM) architecture, which updates recurrent weights to overcome the vanishing gradient problem. This, in turn, improves training performance. Methods: The RNN model is developed based on stack LSTM and bidirectional LSTM. The parameters like mean absolute error (MAE), standard deviation error (SDE), and root mean squared error (RMSE) are utilized as performance measures for comparison with recent state-of-the-art techniques. Results: Results showed that the proposed technique outperformed the existing techniques in terms of RMSE and MAE against all the used wind farm datasets. Whereas, a reduction in SDE is observed for larger wind farm datasets. The proposed RNN approach performed better than the existing models despite fewer parameters. In addition, the approach requires minimum processing power to achieve compatible results.
ABSTRACT
This study evaluated the usability of a direct manipulation device (touchscreen) vs. indirect manipulation devices (mouse and touchpad) on the selected Microsoft (MS) Word tasks as per ISO-9241-11 standard. MS Word was taken as an example of a complex application. The tasks were evaluated in terms of touch-friendly or click-friendly using efficiency, effectiveness, and satisfaction parameters to propose a customized task menu. The experiment was conducted with fifty-four participants, divided into three MS Word usage-based expertise groups. Each participant performed fifty-six tasks using a mouse, a touchpad, and a touchscreen. To assess task-level usability, individual one-way ANOVAs were performed for each task to gauge both efficiency and effectiveness. It's worth noting that the touchscreen significantly outperformed other input methods in just one specific task regarding effectiveness. Consequently, an ANCOVA was employed, with task completion time as the independent variable and the number of errors as a covariate, to further investigate effectiveness. A total of 19 (34%) of the total tasks were found to be significantly efficient with a mouse, while 21 (37.5%) were significantly efficient with a touchscreen. Based on the results, a customized menu is recommended for MS Word-like applications that combine actions in touch-friendly tasks and mouse-friendly tasks separately.
Subject(s)
User-Computer Interface , Humans , Analysis of VarianceABSTRACT
Inductive Teaching Method (ITM) promotes effective learning in technological education (Felder & Silverman, 1988). Students prefer ITM more as it makes the subject easily understandable (Goltermann, 2011). The ITM motivates the students to actively participate in class activities and therefore could be considered a better approach to teach computer programming. There has been little research on implementing ITM in computer science courses despite its potential to improve effective learning. In this research, an existing computer programming lab course is taught using a traditional Deductive Teaching Method (DTM). The course is redesigned and taught by adopting the ITM instead. Furthermore, a comprehensive plan has been devised to deliver the course content in computer labs. The course was evaluated in an experiment consisting of 81 undergraduate students. The students in the Experimental Group (EG) (N = 45) were taught using the redesigned ITM course, whereas the students in the Control Group (CG) (N = 36) were taught using the DTM course. The performance of both groups was compared in terms of the marks obtained by them. A pre-test conducted to compare pre-course mathematical and analytical abilities showed that CG was better in analytical reasoning with no significant differences in mathematical abilities. Three post-tests were used to evaluate the groups theoretical and practical competence in programming and showed EG improved performance with large, medium, and small effect sizes as compared to CG. The results of this research could help computer programming educators to implement inductive strategies that could improve the learning of the computer programming.
