Bulletin of Electrical Engineering and Informatics, 2025
In Ethiopia, where soybeans are mainly involved, manual observation has traditionally been relied... more In Ethiopia, where soybeans are mainly involved, manual observation has traditionally been relied upon for detecting soybean leaf diseases. However, the manual process is susceptible to numerous issues such as laborintensiveness, inconsistency, and subjectivity. While previous studies have explored automated classification for soybean leaf disease detection, they primarily focused on binary classification, overlooking the complexity and diversity of soybean leaf diseases, which hinders effective management strategies. This study introduces deep learning algorithms and computer vision for automated soybean leaf disease identification and classification in soybean leaves. By comparing pre-trained convolutional neural network (CNN) models (VGG16, VGG19, and ResNet50V2), a dataset of 3078 soybean leaf images was curated, representing various diseases. Image preprocessing techniques augmented the dataset to 6,958 images, enhancing the model's accuracy and generalization performance. VGG16 demonstrated outstanding performance with a test accuracy of 99.35%, highlighting its promising performance and generalization potential.
Bulletin of Electrical Engineering and Informatics, 2025
Mobile ad hoc networks (MANETs) operate without fixed infrastructure, with mobile nodes acting as... more Mobile ad hoc networks (MANETs) operate without fixed infrastructure, with mobile nodes acting as both hosts and routers. These networks face challenges due to node mobility and limited resources, causing frequent changes in topology and instability. Clustering is essential to manage this issue. Significant research has been devoted to optimal clustering algorithms to improve cluster-based routing protocols (CBRP), such as the weighted clustering algorithm (WCA), optimal stable clustering algorithm (OSCA), lowest ID (LID) clustering algorithm, and highest connectivity clustering (HCC) algorithm. However, these protocols suffer from high re-clustering frequency and do not adequately account for energy efficiency, leading to network instability and reduced longevity. This work aims to improve the CBRP to create a more stable and long-lasting network. During cluster head (CH) selection, nodes with high residual energy or degree centrality are chosen as CH and backup cluster head (BCH). This approach eliminates the need for re-clustering, as the BCH can seamlessly replace a failing CH, ensuring continuous cluster maintenance. The proposed modified clusterbased routing protocol (MCBRP) evaluated network simulator 2 (ns2) demonstrates that MCBRP is more energy-efficient, selecting optimal CH and balancing the load to enhance network stability and longevity.
Bulletin of Electrical Engineering and Informatics, 2025
The categorization of opinions into positive, negative, or neutral facilitates information gather... more The categorization of opinions into positive, negative, or neutral facilitates information gathering, pinpointing individual weaknesses, and streamlining the decision-making process. Precision in opinion classification enables decision-makers to extract valuable insights, make well-informed decisions, and execute suitable actions. Sentiment analysis is language-specific due to the distinct morphological structures unique to each language, distinguishing them from one another. This study implemented a rule-based sentiment analysis approach for Kafi-noonoo opinionated texts, leveraging a rule-based system tailored for smaller datasets that operate based on a predefined set of rules. The rule-based mechanism calculates the overall polarity of a given sentence by applying a set of rules and categorizes it into positive, negative, or neutral sentiments upon identifying sentimental terms from a dedicated file. While the analysis utilized 1,500 words sourced from Facebook and music review samples, the modest sample size yielded satisfactory results. Performance evaluation metrics such as precision, recall, and F-measure were employed, indicating positive word scores of 91%, 86%, and 88.4%, and negative word scores of 80%, 75%, and 77%, respectively.
Various forms of a single word come to be a big issue for many natural language processing applic... more Various forms of a single word come to be a big issue for many natural language processing applications and related tasks like information retrieval, machine translation, search engine, etc. Especially, for information retrieval unless any mechanism applicable to handle morphological complexes, it degrades retrieval effectiveness (specifically, recall values). By considering this problem, our work focuses on developing a conflation technique called designing a stemmer for Kafi-noonoo text. It aims to design a stemmer for Kafi-noonoo language and the system takes as input a word and removes its affixes according to a rulebased algorithm. This study uses hybrid (longest match and iterative) technique to remove the attached suffixes. Since, there is no prefixation and rare occurrence of infixing in Kafi-noonoo language. Error counting technique was employed to evaluate the performance of this stemmer. For testing purpose, we use 2,075 distinct number of words gathered from different so...
Uploads
Papers by mareye zeleke