Thesis (Master of Science/Engineering in Computer Science and Engineering)
Browse
6 results
Search Results
Item Evidence-based workplace harassment guidance system: transforming #MeToo narratives into actionable knowledge through machine learning(BRAC University, 2026-01) Kabir, Mashphey Bintey; Alam, Md. Golam RabiulWorkplace harassment remains a pervasive global issue, yet victims often lack ac- cess to evidence about what actually happens when people in similar situations take action. Traditional resources offer generic procedural advice without outcome data, leaving individuals to make consequential decisions with incomplete informa- tion. This study presents an evidence-based workplace harassment guidance system that transforms 15,835 #MeToo narratives into personalized, actionable guidance grounded in documented outcomes from comparable cases. The system addresses a fundamental information asymmetry: while organizations accumulate knowledge about harassment cases, individual victims rarely know what outcomes others in similar situations experienced. By analyzing patterns across thousands of documented experiences, the system identifies a user’s specific vulner- ability profile, retrieves semantically similar historical cases, and presents evidence- based guidance including proven successful action sequences, outcome statistics, and high-impact actions that correlate with positive results. The guidance generation pipeline employs SimCSE-BERT embeddings for semantic similarity, multi-label vul- nerability detection leveraging these embeddings to identify seven co-occurring risk factors, and outcome pattern analysis across fifteen outcome categories—supported by BERT sentiment analysis (98.1% accuracy) and HDBSCAN clustering for eval- uation. The empirical analysis reveals sobering realities: negative outcomes predominate across all vulnerability types, institutional inaction occurs in 16.9% of cases, and harasser accountability remains rare at 3.6%. Rather than offering false reassur- ance, the system presents these evidence-based statistics to enable informed decision- making. Professional evaluation with eleven practitioners from HR, legal, mental health, and other sectors—91% with direct harassment case experience—validates that the guidance meets practical standards for appropriateness (M=4.09/5), use- fulness (M=4.09/5), actionability (M=3.91/5), and safety (M=3.73/5), with 91% endorsement and 73% rating the approach superior to typical harassment resources. This research demonstrates that machine learning can provide meaningful support for sensitive domains when developed with rigorous professional validation and com- mitment to user safety.Item Designing a Bangla conversational AI agent for maternal health using model context protocol(BRAC University, 2025-09) Muntasir, Fahim; Noor, JannatunMaternal health in Bangladesh faces persistent challenges, including limited access to skilled providers, informational gaps, and stigma around perinatal mental health. Prior studies show that digital health tools remain constrained by generic content, English-dominated design and neglect of maternal mental health, limiting trust and engagement. We present Baby and Me; the first Model Context Protocol-enabled agentic AI system designed for maternal health in Bangladeshi contexts. The system delivers personalized, empathetic guidance in Bangla and Banglish by combining retrieval-augmented generation with a clinically curated knowledge base, web search, and conversational memory. A survey with 72 women revealed frequent worries about miscarriage, anxiety, and mood changes, highlighting the need for empathetic, accessible support. Evaluation of our prototype showed high contextual accuracy, low hallucination, and strong user satisfaction, with participants valuing empathy and trust while requesting greater personalization. Our findings extend human-AI interaction research by demonstrating how culturally grounded, agentic AI can serve not only as an informational tool but also as a relational companion, offering design insights for equitable health technologies. It paves the way for scalable interventions in low-resource settings, with future directions to enhance maternal outcomes. The chatbot can be accessed in this link.Item Socio-economic impact of 2023 Turkey earthquake price hikes: insightful analysis using transformer models and XAI models(BRAC University, 2024-09) Adib, Muhammed Yaseen Morshed; Sadeque, Farig YousufNatural disasters like the 2023 earthquake in Turkey have significant social and economic effects, making it important to use analytical methods for creating strong, disaster-ready communities. In our work, we analyze public sentiment on social media after the earthquake, focusing on the rise in prices that followed. We classify public reactions into three categories: negative, positive, and neutral. To do this, we use several machine learning models, deep learning models, and two transformer based models. By analyzing the connection between people’s feelings and socio-economic factors like consumer spending, inflation, and price hikes, we aim to understand how public sentiment relates to policy decisions made in response to the crisis. Among all models tested, modified DistilBERT stood out, delivering the best performance with an accuracy of 82.20% and an F1-score of 84.30%. This shows that transformer-based models, particularly DistilBERT, are highly effective for sentiment analysis in this context. DistilBERT’s strong precision, recall, and F1-score suggest that it could be a valuable tool for informing policy changes to reduce the socio-economic impacts of natural disasters. Additionally, we used Explainable AI to help explain the model’s results, ensuring that policymakers can make informed decisions based on the data. Our research highlights the importance of advanced natural language processing (NLP) techniques for developing evidence-based policies in disaster management.Item Curious learner: a generative neuro-symbolic approach for function execution & illustration using natural language(BRAC University, 2024-02) Joaa, A.F.M. Mohimenul; Sadeque, Farig YousufGenerative models possess immense potential, but their ability to perform complex calculations is limited by the need to memorize vast amounts of data, leading to computational inefficiencies. Leveraging tools like the Arithmetic Logic Unit using symbolic functions offers a more efficient alternative, enabling faster responses, smaller model sizes, and improved accuracy. We propose a neuro-symbolic generative model to empower natural language models with task execution abilities by integrating functional programming principles. Experiments on our scoped four translation tasks using 98 mathematical functions demonstrated rapid convergence and minimal training time requirements. Our model, containing 111 million trainable parameters, achieved an average accuracy, BLEU score, and perplexity score of 0.85, 0.84, and 5.9, respectively, after training on a T4 GPU for several hours. This neurosymbolic Language Model shows significant potential for various applications, such as NLP-based command line tools, customer service automation, service discovery automation, project code automation, and natural language-based operating systems.Item A conceptual framework for analyzing critical cactors of PDS learning experience and PDE employment experience(BRAC University, 2023-12) Ahmed, Masum Uddin; Rahman, Mohammad ZahidurThis study aims to examine the relationship and status of education and employment for people with disabilities in Bangladesh. The study employs a mixed-methods approach, including a survey of people with disabilities to grasp their educational and employment situations. The purpose of this study investigation is to explore the factors impacting students and employees with disabilities in Bangladesh. The questionnaire consisted of a mix of numerical, categorical, and multiple-choice questions. This paper adopts multiple data science approaches to measure the reliability between survey items. Seven factors under three dimensions for students with disabilities (PDS) and eight factors under three dimensions for employees with disabilities (PDE) were examined to analyze the influence of their learning and employment experiences. A total of 208 responses were collected from students and employees, and 200 valid responses were retained after data cleaning. Necessary data pre-processing was applied. From the findings, eight factors influencing the learning experiences of students and employment experiences of employees were identified. Finally, the analysis results are presented in the form of suggestions for developing inclusive learning and employment opportunities for individuals with disabilities in Bangladesh. The survey reveals key obstacles for people with disabilities in Bangladesh, including accessibility issues, inadequate accommodations, negative attitudes, and undervaluation in education and employment. It underscores the urgent need for inclusive policies and more research to support their education and employment. The study highlights the requirement for diverse and more effective research methods to comprehend and provide support for individuals with disabilities in Bangladesh.Item Computer assisted learning dictionary for rural students(BRAC University, 2018-05) Rasel, Annajiat Alim; Chakrabarty, Dr. AmitabhaIt is observed that rural students (the target demography) are falling far behind in the countrywide standard tests like SSC and HSC. As high as up to 80% of students from rural areas are failing in national English examinations due to lack of quality and extra-care received by urban students. Training up school and college teachers from one village at a time will take too long to achieve. Furthermore, the extra-care received by urban students like private tuition, the surrounding environment which uses English a lot would be too expensive. It would be almost infeasible to replicate the urban environment in rural areas. English language skills are usually tested in four key areas, listening, reading, writing and speaking. Public examinations in Bangladesh only tests reading and writing skills. This is done in two parts through English 1st paper Examination and English 2nd paper Examination. These parts focus mostly on reading and writing skills respectively. Both of these skills heavily rely on stock vocabulary. Vocabulary stock depends on how much vocabulary students could learn by reading. On the other hand, being able to read requires some vocabulary. Due to limited vocabulary, it becomes difficult for the students to read and write in English. Sentence construction and being able to express in English is a more advanced task compared to reading. This work focuses on building vocabulary for academic reading for the English 1st part examination. The goal is to assist the rural students in acquiring sufficient vocabulary so that they can read the board book prescribed by the National Curriculum & Textbook Board (NCTB) as a Textbook. The more words students know, the more they can read. On the other hand, the more they read the more words they will know. It appears to be a chicken and egg problem. Due to an extremely small size of vocabulary known by students from rural areas of Bangladesh, it becomes tough for them to increase their vocabulary size and to be able to read. It leads to nationwide failure in English examination in addition to Mathematics. This work investigates what approach student may follow to acquire more vocabulary to make their reading smoother. Despite the availability of resources for this purpose, according to our investigation, no empirical research has yet identified the learning environments and the unique learning requirements of the target demography. This work aims to explore the possibilities of automated morphosyntactic tagging of the educational material for these courses and automated extraction of linguistic patterns from the tagged corpora.