http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16360 2026-04-05T17:19:39Z 2026-04-05T17:19:39Z Hasnat, Abul http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16542 2026-03-31T21:01:33Z 2025-01-13T00:00:00Z

An Evaluation of Bangla Text Summarization and Long Context Understanding Using LLMs Hasnat, Abul In the realm of natural language processing, summarizing text in Bangla involves overcoming significant challenges, owing to the language's complex grammatical intricacies and rich linguistic variations. This study explores the effectiveness of state-of-the-art (SOTA) models in summarizing Bangla texts while preserving their essential meaning. The findings reveal that the Gemma series, particularly the Gemma 2 9B model, outperforms the latest Llama series model, Llama 3.1 8B, in capturing the essence of Bangla content, even as context length increases. The Mistral-based Microsoft FILM model, however, emerges as a formidable contender, closely rivaling both Gemma and Llama. Interestingly, despite its advanced architecture, the Llama 3 8B model struggles to match the performance of the Gemma 2B model. Masters of Thesis

2025-01-13T00:00:00Z Mohsin, MD. http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16541 2026-03-31T21:01:32Z 2025-01-18T00:00:00Z

A False Sense Of Security: Enhancing Resilience Against Cyberattack Through An Improved Security Awareness Model Mohsin, MD. Cybersecurity has become a crucial concern in today‘s increasingly digital world, especially for students who rely on online platforms for education, communication, and personal activities. This thesis investigates the effectiveness of university cybersecurity programs in enhancing students‘ online safety practices, with a focus on the integration of both technological and practical applications. Advanced Persistent Threats (APTs) remain a persistent problem within the realm of cybersecurity. Heavy investment effort in security infrastructure and awareness leaves many organizations with a false sense of well-being, for their perceived cybersecurity resilience does not correspond with actual vulnerabilities. This study combines two prior research areas on cybersecurity perception and extends it by implementing Zero Trust Architecture (ZTA) and Artificial Intelligence-Based Threat Detection (AI-TD) tothe current model. Through extensive survey-based investigation, this study evaluates how these new variables impact the confidence towards cyber security in educational and corporate environments, providing valuable solutions to improve the existing cyber security frameworks. Moreover, this research examines the perspective of the faculty, students, IT administrators, and security professionals about the relevance and practical application of these interventions against any cyber threat. Masters of Thesis

2025-01-18T00:00:00Z Zoynob, Bibi http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16540 2026-03-31T21:01:31Z 2025-01-18T00:00:00Z

Statistical Analysis of Post Covid Symptoms on Recovered Patients With ML Zoynob, Bibi As the COVID-19 pandemic elapses, too many attention is being given to long COVID: it is a set of symptoms that people experience after covid infection. Since it is so new, we are still learning about it. The more we understand post-covid syndromes, the more it can help people with both long COVID and everybody else. Those with long COVID is called ‘long haulers’, with the research community it is still settling on a formal term that’s accurate. The study aims to investigate the clinical manifestations and vaccination which were received by the patients that, examining that the patients has post Covid syndrome even if they received Covid vaccine or not. By integrating insights from diverse fields, this thesis endeavors to provide a holistic understanding of Post Covid Syndrome’s and foster targeted interventions for patient care and health management. So that, patients can be awarded about taking vaccination. Covid-19 vaccinations currently authorized for Bangladesh are Pfizer and Moderna. This thesis research will try to resolve the complexity of post Covid syndrome for these vaccination in Bangladesh. Masters of Thesis

2025-01-18T00:00:00Z Rahman, Sheikh Md. Mizanur http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16539 2026-03-31T21:01:29Z 2025-02-03T00:00:00Z

Online Order System App for Pharmaceiticals Rahman, Sheikh Md. Mizanur The rapid advancement of digital technology has significantly transformed the healthcare and pharmaceutical industries. This thesis presents the development of an "Online Order System App for Pharmaceuticals," specifically tailored for Beacon Pharmaceuticals, to streamline and enhance operational efficiency. The app serves as a centralized platform for field medical information officers (MIOs) to manage essential tasks, including ordering products for customers and patients, accessing product pricing and offer details, and generating invoices. Furthermore, the system incorporates delivery tracking features, enabling real-time updates on order statuses. To support pharmaceutical sales and marketing efforts, the app facilitates the submission of doctor visit reports and doctor prescription (RX) reports, ensuring seamless communication and record-keeping. Additionally, the inclusion of sales target and achievement tracking provides MIOs and management with a comprehensive overview of performance metrics. By integrating these functionalities, the proposed app addresses challenges such as manual data entry, fragmented communication, and inefficiencies in order management. The study also highlights the design process, implementation strategies, and the potential impact of this system on operational workflows. Ultimately, this research underscores the value of digital solutions in optimizing pharmaceutical operations and enhancing customer satisfaction. Masters of Thesis

2025-02-03T00:00:00Z