IJMECS Vol. 14, No. 3, Jun. 2022
Cover page and Table of Contents: PDF (size: 606KB)
During the recent Covid-19 pandemic, there has been a tremendous increase in online-based learning (e-learning) activities as nearly every educational institution has transferred its programs to digital platforms. This makes it crucial to investigate student performance under this new mode of delivery. This research conducts a comparison among the traditional educational data mining techniques to detect the best performing classifier for analyzing as well as predicting students’ performance in online learning platforms during the pandemic. It is achieved through extracting four datasets from X-University student information system and learning platform, followed by the application of 6 classifiers to the extracted datasets. Random Forest Classifier has demonstrated the highest accuracy in the first two out of the four datasets, while Simple Cart and Naïve Bayes Classifiers presented the same for the remainder two. All the classifiers have demonstrated medium to high TP rates, class precision and recall, ranging from 60% to 100% for almost all of the classes. This study emphasized the attributes that have a direct impact on students’ performance. The outcomes of this study will assist the instructors and educational institutions to identify important factors in the analysis and prediction of student performance for online program delivery.[...] Read more.
Software development agility has been regarded as a critical pillar of modern businesses. However, there is still a way to find whether there exists a consistent, complete, precise, agreed and uniformed definition of it. In this regard, this study firstly reviews the existing definitions of agility in the software development domain from the literature. As one of the main results of this phase, we have seen that although agility has a remarkable root in the software development domain, even its definition is still debatable and there are other concepts close to agility in terms of definition but used interchangeably. There is another confusion about how some researchers define agility over other different concepts, although there is no apparent unifying factor in their origins except their historical co-occurrence. In addition, there are particular practices embedded into the agility definitions mostly from the manifesto and Scrum. After uncovering the deficiencies of the existing definitions, we aimed to ratify the definition of the agility concept. Then, we intended to synthesize the underlying facets of the identified definitions and propose a new yet more comprehensive definition revealing the agility characteristics properly by considering the interpretations of the existing definitions. Our study stands out by using a customized synthesis method for analysis, providing inputs to this analysis with a comprehensive literature review, and the comprehensive evaluation of the facets with the support of the literature. We are aware that agreeing on a definition is a valuable exercise and a good starting point for a better understanding of the agility phenomenon that could enable and lead to more realistic implementations, less disappointment and disillusionment, and possibly greater success rates for both practitioners and researchers.[...] Read more.
The Recurrent Neural Network (RNN) is well suited for emotional speech recognition because its uses constantly time shifting property. Even though RNN gives better results GRU, LSTM and BILSTM solves the gradient problem and overfitting problem joins the path to reduces the efficiency. Hence in this paper five deep learning architecture is designed in order to overcome the major issues using data augmentation and spatial feature. Five different architectures like: Enhanced Deep Hierarchal LSTM & GRU (EDHLG), EDHBG, EDHGL, EDHGB & EDHGG are developed with dropout layers. The raw data learned from LSTM will be given as the input to GRU layer for deepest learning. Thus, the gradient problem is reduced, and accuracy of each emotion was increased. Also, to enhance the accuracy level spatial features were concatenated with MFCC. Thus, in all models, the experimental evaluation with the Tamil emotional dataset yielded the best results. EDHLG has a 93.12% accuracy, EDHGL has a 92.56 percent accuracy, EDHBG has a 95.42 percent accuracy, EDHGB has a 96 percent accuracy, and EDHGG has a 94 percent accuracy. Furthermore, the average accuracy rate of a single individual LSTM layer is 74%, while BILSTM is 77%. EDHGB outperforms almost all other systems, by an optimal system of 94.27 percent and then a maximum overall accuracy of 95.99 percent. For the Tamil emotion data, emotional states such as happy, fearful, angry, sad, and neutral have a 100% prediction accuracy, while disgust has a 94 percent efficiency rate and boredom has an 82 percent accuracy rate. Also, the training time and evaluation time utilized by EDHGB is 4.43 mins and 0.42 mins which is less when compared with other models. Hence by changing the LSTM, BILSTM and GRU layers large analysis of experiment on Tamil dataset is done and EDHGB is superior to other models, and when compared with basic models LSTM and BILSTM around 26% more efficiency is gained.[...] Read more.
An electromyography (EMG) is an analytical tool used to record muscles' electrical activity, which produces an electrical signal proportional to the level of muscle activity. EMG signal plays a vital role in bio-mechatronic engineering for designing intelligent prostheses and other rehabilitation devices. Analysis of EMG signals with powerful and advanced methodologies is an essential requirement in EMG signal processing, as the EMG signal is a complex nonlinear, non-stationary signal in nature. It is required to use advanced signal processing techniques rather than conventional methods to exact EMG signals' features. Fourier transforms (FT) are not the most appropriate tool for analyzing non-stationary signals such as EMG. In this work, we have developed a system that can be useful for disabled persons to get a regular lifestyle using a functioning part of the body. Here, we studied the electrocution gram behavior of human body parts to feature extraction and trained the neural network to simulate the movements of mechanical actuators such as robotic arms. The wavelet transformation has been used to get high-quality feature extraction from electro cardio grapy and develops proper faltering methods for cardio systems' electrical signals. Finally, an artificial neural network (ANN) is used to classify the EMG signals through exacted features. Classification results are presented in this paper.[...] Read more.
The purpose of this study is to conduct an empirical investigation and comparison of the effectiveness of various classifiers and ensembles of classifiers in predicting academic performance. The study will evaluate the performance and efficiency of ensemble techniques that employ several classifiers against the performance and efficiency of a single classifier. Reducing student attrition is a serious concern for educational institutions worldwide. Educators are looking for strategies to boost student retention and graduation rates. This is only achievable if at-risk students are appropriately recognized early on. However, most commonly used predictive models are inefficient and inaccurate due to intrinsic classifier limitations and the usage of minor factors. The study contributes to the body of knowledge by proposing the development of optimized ensemble learning model that can be used for improving academic performance prediction. Overall, the findings demonstrate that the approach of employing optimized ensemble learning (OEL) model approaches is extremely efficient and accurate in terms of predicting student performance and aiding in the identification of students who are in the fear of attrition.[...] Read more.
Autonomous driving is predicted to alter the transportation industry in the near future. For decades, carmakers, researchers, and administrators have already been working in this sector, with tremendous development. Nevertheless, there are still many uncertainties and obstacles to solve, not only in terms of technical technology, as well as in terms of human consciousness, culture, and present traffic infrastructure. With respect to technological challenges, precise route identification, avoiding the improper location, time delay, erroneous drop-off, unsafe path, and automated navigation in the environment are only a few. RL (Reinforcement Learning) has evolved into a robust learning model which can learn about complications in high dimensional settings, owing to the advent of deep representation learning. Environment learning has been shown to reduce the required time delay, reduce cost of travel, and improve the performance of the agent by discovering a successful drop-off. The major goal is to ensure that an autonomous vehicle driving can reach passengers, pick them up, and transport them to drop-off points as quickly as possible. For performing this task, RL methods like DQNs (Deep Q Networks), Q-LNs (Q-Learning networks) , SARSAs (state action reward state actions), and ConvDQNs (convolution DQNs) are proposed for driving Taxis autonomously. RL agent’s decisions are based on MDPs (Markov Decision Processes). The agent has effectively learnt the closest path, safety, and lower cost, gradually obtaining the capacity to travel bigger areas of the successful drop-off without negative incentive for reaching the target using these RL approaches. This scenario was chosen based on a set of requirements for simulating autonomous vehicles using RL algorithms. Results indicate that ConvDQNs are capable of successfully controlling cars in simulation environments than other RL methods. ConvDQNs are a combinations of CNNs (Convolution Neural Networks) and DQNs. These networks show better results than other methods as their combining of procedures gives improved results. Results indicate that ConvDQNs are capable of successfully controlling a car to navigate around a Taxi-v2 environment than the existing RL methods.[...] Read more.