BRAC University Institutional Repository
Preserving Knowledge, Advancing Research, Sharing Scholarship
A digital platform for collecting, preserving, and sharing BRAC University’s scholarly, academic, and institutional outputs.
Contribute to the Repository
Faculty members and students are invited to submit their research publications, theses, dissertations, and scholarly works to increase visibility, access, and long-term preservation.
Recent Submissions
Economic load dispatch using the TLBO algorithm
(Institute of Electrical and Electronics Engineers Inc., 2024-01-01) Fahim K.E.; Tabassum T.; Islam, Rakibul; Rahman, Maria; Jonayed, Khondaker Labib Al; Faiyaz, Fahim; Department of Electrical and Electronic Engineering
The Teaching-Learning-Based Optimization (TLBO) algorithm is suggested in this paper as a potential solution to the economic load dispatch (ELD) problem in power systems. ELD, a fundamental issue in power systems, tries to schedule generator power production in a way that satisfies load demand while reducing operating costs. The teaching and learning processes in a classroom served as the inspiration for TLBO, a relatively recent optimization algorithm that has demonstrated promising results in handling challenging optimization issues. The suggested approach is evaluated using a variety of power systems with various characteristics, and the outcomes are contrasted with those of other cutting-edge optimization algorithms. The contribution of this paper lies in assessing the suitability of this algorithm for solving real-world economic load dispatch problems and contrasting it with industry-standard software simulation results, such as the Power World Simulator.
SmartCitrus: An efficient deep learning approach for real-time detection and classification of citrus leaf diseases
(Institute of Electrical and Electronics Engineers Inc., 2024-01-01) Emon, Shaharear Hossain; Islam, Iftea Khairul; Nahin, Tasfia Jahan; Ahmed, Ahnaf Mahdin; Orchi, Nabiha Tasnim; Alam, Md Ashraful; Dipto, Shakib Mahmud; Department of Computer Science and Engineering
Bangladesh is a prominent citrus exporter. Annually, the country has been exporting citrus fruits to over 60 countries. Distinguishing various diseases affecting citrus leaves requires a significant investment of time, effort, and specialized knowledge. Consequently, it is essential to create an innovative method for detecting citrus diseases. In this study, we have devised a valuable methodology by employing CNN models to identify diseases in citrus leaves. By employing a distinctive ensemble strategy, we successfully trained the model using varying numbers of classes in each stage. In reality, it allowed us to utilize suitable varieties of leaves for various ailments. Furthermore, it has enhanced the rate at which models learn during the later stages. In addition, it has reduced the level of model intricacy in comparison to frequently employed ensemble models. The identification of plant diseases in the present study involved the utilization of leaf photographs and algorithms for segmentation and feature extraction. Ultimately, we have successfully attained a 96 % accuracy rate for each class, signifying a substantial potential for mitigating production losses.
Enhancing transparency in transport mode detection: An interpretable ensemble model classifier
(Institute of Electrical and Electronics Engineers Inc., 2024-01-01) Aziz, Azwad; Youkee, Nafisa Khan; Ahmed, Fariha Shams; Tasnim, Sandia; Fahim-Ul-Islam, Md.; Chakrabarty, Amitabha; Department of Computer Science and Engineering
Transport Mode Detection (TMD) is a crucial component in the field of Intelligent Transportation Systems (ITS), taking advantage of current advancements in Artificial Intelligence and the Internet of Things (IoT). This research undertakes a thorough investigation of transportation modalities, recognizing the crucial role of TMD in increasing road safety. The study evaluates established machine learning methodologies, such as LightGBM, XGBoost, and CatBoost. Furthermore, it introduces a novel ensemble model that capitalizes on the advantages of these techniques, resulting in higher precision when categorizing transportation modes. Notably, the ensemble model provides accuracies of 83%, 93%, and 94% across three independent sensor data collections from the TMD dataset, surpassing the performance of alternative machine learning classifiers. In addition, the incorporation of Shapley Additive Explanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) improves the comprehensibility of the model we propose, providing a valuable understanding of the decision-making procedures. This research not only enhances the current endeavors to improve road safety but also presents a promising method for the advancement of intelligent transportation systems.
Enhancing precision in rice leaf disease detection: A transformer model approach with attention mapping
(Institute of Electrical and Electronics Engineers Inc., 2024-01-01) Ahmed, Sarder Tanvir; Barua, Shomtirtha; Fahim-Ul-Islam, Md.; Chakrabarty, Amitabha; Department of Computer Science and Engineering
Advancements in image recognition technology have significantly impacted agricultural practices, especially in early detection of rice leaf diseases, which is crucial for maintaining crop health and yield. This paper introduces the Optimized BEiT model, a novel lightweight CNN and transformer architecture, specifically designed for this purpose. The model demonstrates high accuracy, outperforming traditional models with a precision of 0.92, recall of 0.91, and F1-score of 0.91. It was rigorously trained and validated on a comprehensive dataset featuring healthy and unhealthy rice leaf images. The use of attention mapping techniques such as GRAD-CAM has been pivotal in interpreting the model's predictions, enhancing the transparency and reliability of AI in agricultural diagnostics. These techniques provide clear, comprehensible insights into the predictive features of the model, ensuring its decisions are understandable and trustworthy. This research is a significant stride in precision agriculture, offering a robust tool for agricultural professionals. This model not only represents a technical achievement but also a practical solution for real-world agricultural challenges, demonstrating the potential of AI to enhance food security and sustainable farming practices.
Decoding emotions: Leveraging machine learning to analyze emotional expression for deeper mental health insights
(Institute of Electrical and Electronics Engineers Inc., 2024-01-01) Wahiduzzaman, Md.; Fahim-Ul-Islam, Md.; Chakrabarty, Amitabha; Department of Computer Science and Engineering
The increasing frequency of mental health illnesses, their multiple symptoms, and complications with other conditions bring complexity to diagnosis and increase the risk of misdiag-nosis. This research emphasizes the growing need for healthcare practitioners to evaluate patients' mental health memories for more effective therapy. The World Health Organization's (WHO) research underlines the severity of mental health problems as a major contributor to suicide, particularly affecting the young population. Stress, frequently unrecognized, can have negative repercussions, needing early identification and intervention. With mental health disorders known as a substantial contributor to sui-cide, early detection is vital. Early identification of mental health disorders, is essential to prevent the worst outcome, decrease the risk of suicide, boost the effectiveness of therapy, improve overall well-being, and provide cost-efficient therapies. Therefore, leveraging automated classifiers, including SVM, BERT, Random Forest, Logistic Regression, and a Proposed CNN model, the study evaluates emotional expression in online posts by cancer patients. The Proposed CNN model surpasses existing classifiers in classification and sentiment analysis with an accuracy of 89.12%, emphasizing the transformative potential of machine learning (ML) in mental health research for early identification and improving the treatment process.