http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16 Fri, 26 Jun 2026 11:17:35 GMT 2026-06-26T11:17:35Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17559 Reasoning Over Context in Bangla: A Generative QA Approach to Factoid Understanding Using LLM Rana, Md. Masud This paper presents the development of a Bangla Question Answering (QA) system using advanced transformer-based models to tackle the complexities of Bangla language processing. Specifically, it compares the performance of BanglaT5, a model fine-tuned for Bangla, with mT5, a multilingual variant of the T5 model. Both models were evaluated on a dataset of over 7,500 Bangla news articles, focusing on factoid-based question answering. The results show that BanglaT5 outperforms mT5 on key metrics such as ROUGE, BLEU, Character Error Rate (CER), and Word Error Rate (WER), showcasing its superior ability to handle Bangla’s unique linguistic features like morphology and syntax. BanglaT5 achieved a ROUGE-1 F1 score of 0.6979, Exact Match Accuracy of 0.49, and CER of 0.4054, demonstrating its ability to generate accurate, contextual answers. In contrast, mT5’s performance was much lower, with an Exact Match Accuracy of 0.0008 and WER of 0.9996. This comparison highlights the importance of fine-tuning models for specific languages like Bangla, emphasizing the limitations of multilingual models in tasks requiring deep linguistic understanding. The system developed in this research offers a scalable solution for Bangla QA, with potential applications in education, public services, and digital literacy, contributing to the growing field of Bangla NLP. Future work will focus on deploying the model in real time, expanding the dataset, and exploring multimodal capabilities to increase its use in real-world applications. Project Report Tue, 14 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17559 2025-01-14T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17558 Ideation of Depression and Suicide Using Machine Learning Techniques Rihan, Aspy Depression has emerged as a major mental health challenge globally, with a noticeable rise in prevalence in Bangladesh, particularly accompanied by increasing suicidal tendencies. This study investigates the underlying causes of depression and presents a machine learning-based approach for its early detection. Unemployment, family pressure, work stress, and social isolation were identified as key contributing factors. This issue was addressed using several supervised machine learning models, including Nave Bayes, Support Vector Machine (SVM), Logistic Regression (LR), Random Forest, Linear Discriminant Analysis (LDA), AdaBoost, Decision Tree, and k-Nearest Neighbor (k-NN). A comprehensive dataset related to mental health and depression symptoms was used to train and test these models. Statistical Machine Learning has shown to be the most accurate and consistent method while Logistic Regression provided the most consistent and balanced performance. Naïve Bayes had good recall capabilities, and AdaBoost had robust performance across a variety of metrics. Additionally, Random Forest and k-NN provided reliable results, while Decision Tree and LDA did not produce any interpretable yet effective results. This study confirms the potential of machine learning techniques for the accurate detection of depression and related mental health issues. It will be important to enhance model explanation, reduce algorithmic bias, integrate diverse data sources, and adhere to ethical principles like privacy protection and informed consent for future research. In resource-constrained regions like Bangladesh, AI-driven mental health tools are especially important for enabling timely diagnosis and support. Project Report Tue, 14 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17558 2025-01-14T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17556 Identifying The Authenticity of Images Using Deep Learning Techniques Mondal, Sourav Kumar The rapid development of generative artificial intelligence has produced a wave of hyper-realistic deepfakes, posing existential challenges for authenticity verification to occur in digital media. This study proposes a deep learning architecture for the binary classification of AI-generated and real images as a reaction to a growing need for credible detection techniques. We comparatively evaluate four various architectures: ResNetRS50, MobileNetV2, EfficientNetB0, and a specially designed CNN with integrated Gabor filters and attention mechanisms. All models were trained and evaluated on an equalized, high-quality dataset under the same experimental conditions to provide serious benchmarking. While MobileNetV2 and EfficientNetB0 achieved higher peak validation accuracies of 99.29% and 99.81% respectively, ResNetRS50 was the most powerful and most generalized model. Its robust convergence behavior, high interpretability, and resistance to overfitting— particularly under extended training durations and high-density data—make it the top choice even at a slightly lower peak accuracy of 97.24%. Extended testing using classification reports, confusion matrices, and performance curves supports this conclusion further. A web interface was also established to demonstrate real-time deployment capability, showing that the model is usable in practical applications. The proposed method not only elevates the state of AI image forensics but also serves as a basis for large-scale and trustworthy content verification systems in the face of rising synthetic media. Project Report Tue, 14 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17556 2025-01-14T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17551 Fruit Quality Classification using Deep Learning and Explainable AI Annafi, Abdun Nafi; Muhtasim, Tasnuva Fruit quality classification is a critical task in agriculture and food industries to ensure standardization and market value. This study evaluates the performance of VGG19, MobileNetV2, ResNet50, custom CNN and BiLSTM for fruit classification using deep learning. A dataset of 3,758 images across seven fruit classes was used for training and evaluation. Among the tested models, MobileNetV2 achieved the highest accuracy (99.48%), making it the most suitable for real-world applications due to its efficiency. LIME (Local Interpretable Model-Agnostic Explanations) was employed to interpret model predictions, verifying that fruit characteristics like color, shape, and texture were key factors in classification decisions. The study highlights dataset imbalance and lighting variations as primary challenges. Future improvements include dataset expansion, hyperparameter optimization, and real-time deployment of the best-performing model. This research provides insights into selecting optimal deep learning models for automated fruit classification, contributing to precision agriculture and quality assurance in food industries. Project Report Tue, 14 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/17551 2025-01-14T00:00:00Z