Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Link stability based multipath routing and effective mobility prediction in cognitive radio enabled vehicular ad hoc network
Get PDFVehicular ad hoc networks (VANETs) provide a robust infrastructure for intelligent transportation system (ITS) applications. VANET communication involves vehicle-to-vehicle and vehicle-to-infrastructure connections, primarily with roadside units (RSUs). Analyzing cognitive radio (CR)-VANET studies revealed two key performance issues: high energy consumption and latency. To address these challenges, we propose a novel approach: link stability and mobility prediction-based clustered CR-VANETs, known as LMCCR-VANET. LMCCR-VANET consists of four main components: CR-VANET construction, clustering model, speed-based mobility prediction, and link-based multipath routing. Initially, we establish cluster-based CR-VANETs to analyze and mitigate spectrum scarcity and power utilization problems in VANETs. Mobility prediction evaluates vehicle speed variations and predictions. Finally, employing link stability-based multipath routing (LSMR) in conjunction with the fuzzy interference model and ad hoc on-demand multipath distance vector (AOMDV) routing protocol ensures stable and efficient routing. Experimental results showcase the superiority of LMCCR-VANET. It exhibits enhanced energy efficiency, delivery rates, reduced energy consumption, end-to-end latency, and routing overhead when compared to recent works such as SCCR-VANET, CFCR-VANET, and MMCR-VANET
Sentiment analysis of imbalanced Arabic data using sampling techniques and classification algorithms
Get PDFSentiment analysis is a popular natural language processing task that recognizes the opinions or feelings of a piece of text. Microblogging platforms such as Twitter are a valuable resource for finding such people’s opinions. The majority of Arabic sentiment analysis studies indicated that the data utilized to train machine learning algorithms is balanced. In this paper, we investigated the impact of sampling techniques and classification algorithms on an imbalanced Arabic dataset about people’s perceptions of COVID-19, with the majority of opinions reflecting people’s fear and stress about the pandemic, and the minority reflecting the belief that the pandemic was a hoax. The experiments concentrated on analyzing the imbalanced learning of Arabic sentiments using over-sampling and under-sampling techniques on seven single machine learning algorithms and two common ensemble algorithms from the bagging and boosting families, respectively. Results show that resampling-based approaches can overcome the difficulty of an imbalanced dataset, and the use of over-sampled data leads to better performance than that of under-sampled data. The results also reveal that using oversampled data from synthetic minority over-sampling technique (SMOTE), borderline-SMOTE, or adaptive synthetic sampling with random forest classifier is the most effective in addressing this classification problem, with F1-score value of 0.99
A novel agile method for user stories’ XMI model generation via NLP and MDA
Get PDFAgile software development methodologies have grown in popularity during the past few years. One of the key components of agile development is the use of user stories to describe software requirements. However, creating and managing user stories can be time-consuming and error-prone. In this paper, we present a novel method to generating user stories’ XMI model using natural language processing (NLP) and model-driven architecture (MDA) approach. We devel-oped a method that uses NLP to extract key information from user stories and then applies MDA techniques to generate an XMI model conforming to its pro-posed meta-model. We conducted a case study to illustrate and validate our method, and we analyze and discuss the studied-related work with our proposal. As a result, our method has the potential to make user stories’ models and their meta-models the focus of software development. This will help to streamline the development process by making it easier to construct and transform models in an agile environment with the MDA approach
Classification of clove types using convolution neural network algorithm with optimizing hyperparamters
Get PDFThis study uses clove imagery by classifying it according to ISO 2254-2004 standards: whole, headless, and mother clove. This type of clove will affect the quality and economic value when it has been dried. For this reason, it is necessary to take a first step to control cloves' quality. One way is to classify it from the start. This research will utilize the convolution neural network algorithm and compare it with model transfer learning and modified VGG16 architecture on clove images. In addition, research is also looking for the most optimal hyperparameter. The results of this study indicate that the application of convolution neural network (CNN) to clove images obtains an accuracy value of 84% using a hyperparameter of 50 epochs, a learning rate of 0.001, and a batch size of 16. Meanwhile, for the application of transfer learning VGG16, Resnet50, MobileNetV2, InceptionV3, DensetNet151, and modified VGG16 have respectively each of the highest accuracy including 95.70%, 76.15%, 96.89%, 98.07%, 98.96%, and 99.11%
Enhance sentiment analysis in big data tourism using hybrid lexicon and active learning support vector machine
Get PDFSentiment analysis is a review analysis process used to determine whether an opinion is neutral, negative, or positive. Sentiment analysis can be done using lexicon-based or machine learning-based approaches. Lexicon can perform sentiment analysis without training data because it is dictionary-based but performs worse than machine learning. Machine learning can perform well in completing sentiment analysis but requires training data so that the model does not experience underfitting. In the case of sentiment analysis on big data, manual labeling of training data is an inefficient job. Support vector machine (SVM) has the opportunity to be used together with the active learning (AL) method to make small training data but still have good performance. This research proposed a hybrid lexicon and AL-SVM method to complete sentiment analysis on big data tourism. This research used polarity from the valence aware dictionary and sentiment reasoner (VADER) lexicon as a reference for the query by user process from the AL-SVM to automate the sentiment analysis process on big data. The experimental results showed that using the hybrid lexicon and AL-SVM increased the sentiment analysis performance compared to the VADER lexicon, SVM, and lexicon SVM, which run separately
Applications of nanostructured materials for severe acute respiratory syndrome-CoV-2 diagnostic
Get PDFThere is a growing concern that severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) infections will continue to rise, and there is now no safe and effective vaccination available to prevent a pandemic. This has increased the need for rapid, sensitive, and highly selective diagnostic techniques for coronavirus disease (COVID-19) detection to levels never seen before. Researchers are now looking at other biosensing techniques that may be able to detect the COVID-19 infection and stop its spread. According to high sensitivity, and selectivity that could provide real-time results at a reasonable cost, nanomaterial show great promise for quick coronavirus detection. In order to better comprehend the rapid course of the infection and administer more effective treatments, these diagnostic methods can be used for widespread COVID-19 identification. This article summarises the current state of research into nanomaterial-based biosensors for quick SARS‑CoV‑2 diagnosis as well as the prospects for future advancement in this field. This research will be very useful during the COVID-19 epidemic in terms of establishing rules for designing nanostructure materials to deal with the outbreak. In order to predict the spread of the SARS-CoV-2 virus, we investigate the advantages of using nano-structure material and its biosensing applications
A review of deep learning models (U-Net architectures) for segmenting brain tumors
Get PDFHighly accurate tumor segmentation and classification are required to treat the brain tumor appropriately. Brain tumor segmentation (BTS) approaches can be categorized into manual, semi-automated, and full-automated. The deep learning (DL) approach has been broadly deployed to automate tumor segmentation in therapy, treatment planning, and diagnosing evaluation. It is mainly based on the U-Net model that has recently attained state-of-the-art performances for multimodal BTS. This paper demonstrates a literature review for BTS using U-Net models. Additionally, it represents a common way to design a novel U-Net model for segmenting brain tumors. The steps of this DL way are described to obtain the required model. They include gathering the dataset, pre-processing, augmenting the images (optional), designing/selecting the model architecture, and applying transfer learning (optional). The model architecture and the performance accuracy are the two most important metrics used to review the works of literature. This review concluded that the model accuracy is proportional to its architectural complexity, and the future challenge is to obtain higher accuracy with low-complexity architecture. Challenges, alternatives, and future trends are also presented
Virtual teaching and learning for autistic students amidst the pandemic: a systematic literature review
Get PDFTeaching and learning for autistic students during the COVID-19 pandemic pose challenges for educators. This systematic literature review (SLR) aimed to explore the effectiveness of virtual teaching and learning (VTL) by employing the reporting standards for systematic evidence syntheses (ROSES) framework. Articles from databases like Scopus, Web of Science, and Google Scholar were systematically examined, focusing on themes such as support, coping strategies, teaching methods, flexibility, and communication. The review identified 14 sub-themes within these categories, providing tailored coping and teaching strategies for parents, teachers, and caregivers working with autistic students. From 706 initially identified articles, 376 were selected, with 17 specifically relevant to virtual teaching for autistic students during the pandemic. These findings contribute insights to the existing literature and offer practical implications to enhance VTL experiences for autistic students facing pandemic challenges
Deep learning based detection, classification, and location of power system faults
Get PDFThe identification, categorization, and localization of faults play a crucial role in maintaining the smooth operation of power systems. Distance relays possess a significant capability to withstand power fluctuations, thereby minimizing inadvertent disruptions in transmission lines. Addressing these challenges involves the adoption of advanced fault analysis techniques to enhance the accuracy and speed of relay operations. While modern machine learning (ML) approaches are still nascent in fault analysis, the authors propose a novel deep learning (DL) based long short term memory (LSTM) method for precise fault detection, classification, and rapid fault location estimation. The proposed approach is applied to the Kundur two-area 4 machine 11 bus system covering a distance of 220 km. The LSTM fault detection (LSTM (FD)) module accurately detects and classifies faults, while the LSTM fault location (LSTM (FL)) module precisely estimates fault locations. The effectiveness of the proposed method is verified through a comparative assessment with various traditional ML and DL techniques. The protection modules are also tested under different fault locations, fault resistances, and noisy signals. The features taken into consideration for the operation of the protection modules are different bus voltages, bus currents, zero sequence voltage, zero sequence current, fault inception angle, and fault resistance
Understanding explainable artificial intelligence techniques: a comparative analysis for practical application
Get PDFExplainable artificial intelligence (XAI) uses artificial intelligence (AI) tools and techniques to build interpretability in black-box algorithms. XAI methods are classified based on their purpose (pre-model, in-model, and post-model), scope (local or global), and usability (model-agnostic and model-specific). XAI methods and techniques were summarized in this paper with real-life examples of XAI applications. Local interpretable model-agnostic explanations (LIME) and shapley additive explanations (SHAP) methods were applied to the moral dataset to compare the performance outcomes of these two methods. Through this study, it was found that XAI algorithms can be custom-built for enhanced model-specific explanations. There are several limitations to using only one method of XAI and a combination of techniques gives complete insight for all stakeholders