http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16554 Mon, 27 Apr 2026 00:31:22 GMT 2026-04-27T00:31:22Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16820 Impact Of Climate Change Disasters on the Educational Sector Of Coastal Area, Bangladesh Almamun, Md. Abdulla Moheshkhali, situated in southeastern Bangladesh, is highly vulnerable to natural disasters, especially cyclones and saline water surges intensified by rising temperatures due to climate change. These events contribute significantly to river erosion, property loss, heightened poverty, and various socioeconomic challenges. This study aims to evaluate how climate change-induced natural disasters impact the district's education sector, causing difficulties in resource allocation, infrastructure maintenance, and resulting in issues such as student dropout, early marriage, and livelihood shifts. Despite extensive research on climate change impacts in Bangladesh's coastal regions, there is a notable lack of publications specifically addressing these issues in Moheshkhali. The study utilizes content analysis to examine relevant books, articles, journal publications, and online media features. Coding of the content analysis is meticulously aligned with the research questions to comprehensively explore causes and effects. The findings underscore climate change's profound influence on student dropout rates, driven by infrastructure damage, population displacement, increased poverty, early marriages, and changes in livelihoods. Addressing these challenges necessitates significant intervention, capacity building, and community adaptation efforts. Project Wed, 15 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16820 2025-01-15T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16819 From Bangladesh to Europe: Evaluating The Impact Of Erasmus Mundus Joint Master’s Degree (Emjm) On Education and Career Development Snigdha, Shouvik Roy The objective of this analysis is to comprehensively evaluate the impact of the Erasmus Mundus Joint Master’s Degree (EMJM) program on the educational and professional outcomes of Bangladeshi participants. This study aims to determine the annual number of scholarships awarded to Bangladeshi students, identifying trends and assessing the accessibility of the program for applicants from the region. Additionally, it seeks to examine the job placement ratio of EMJM graduates, focusing on their integration into various industries and sectors, both in Bangladesh and abroad, and highlighting the program’s contribution to enhancing their employability and career progression. A critical aspect of the analysis involves exploring the broader impact of the program, including the development of academic skills, intercultural competencies, and global networks, while also considering its influence on personal growth. Furthermore, the study aims to assess the retention and return ratio, analyzing the proportion of Bangladeshi graduates who choose to stay in Europe versus those who return home, and exploring the motivations and challenges underlying these decisions. By addressing these dimensions, the analysis intends to provide actionable insights for stakeholders, including policymakers and academic institutions, to optimize the program's benefits for Bangladeshi students, foster international collaboration, and align the outcomes with national development priorities. Project Tue, 14 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16819 2025-01-14T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16817 Bullying Detection from Tweets with LSTM And Bert Transformer Mia, Robin In the internet age, cyberbullying has grown to be a serious concern, particularly for English-speaking nations. This work focuses on identifying cyberbullying in English with the use of deep learning techniques. A specialized English dataset, comprising instances of both cyberbullying and non-cyberbullying text, is utilized for training a deep learning model. Tokenization, preprocessing, and sequence transformation are applied to the dataset so that it may be fed into Random Forest, Naïve Bayes, and BERT classifiers using LSTM cells. The novel LSTM-based deep learning model was used for the dataset and the dropout and word embedding technique were used to improve the model’s performance. The best model was evaluated with confusion matrix. Research is being done on a number of approaches, including language-specific preprocessing and data augmentation, to address the particular problems with cyberbullying detection in English. The results demonstrate how well deep learning works to identify cyberbullying in English-speaking contexts and show how promising the technology is for addressing this issue. The study reveals that the BERT achieved an accuracy of 87%, demonstrating its superior performance. Additionally, an alternative approach using LSTM yielded the accuracy 84%. Ensemble models, including Naïve Bayes (NB), and Random Forest, were also employed, with hyper- parameter tuning optimizing their performance. Notably, the LSTM and BERT outperformed other models, attaining the highest accuracy rate of 87% in cyberbullying detection, as confirmed by recent experimental inquiries evaluating these findings. Project Wed, 15 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16817 2025-01-15T00:00:00Z http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16815 Leveraging Machine Learning for Predictive Analysis Of Drug Addiction in Bangladesh Akter, Fahmida Drug addiction has become a critical health concern, especially among the youth of Bangladesh. According to AsiaNews, there are more than 8 million drug addicts, out of which a large portion consists of young people. Due to this growing menace, some effective preventive steps need to be taken. This research examines some important factors that differentiate between addicted and non-addicted people to help in targeted interventions. This study is based on 1,624 individual responses, collected through an online survey containing a wide array of demographic and behavioral attributes. The dataset includes participants from Dhaka and Sylhet, aged 15 to 27 years, predominantly students. In this paper, a machine learning-based approach was followed for analyzing and predicting the risk of addiction. Models used in this work are Multilayer Perceptron (MLP), Extreme Learning Machine (ELM), CatBoost, and standard classifiers. Ensemble techniques were also adopted, namely Blending (Neural Networks, Gradient Boosting, and K-Nearest Neighbors) and Voting (Logistic Regression, Random Forest, Support Vector Machine). Models were evaluated as well, and their performances corresponded to 98.15% for Blending and CatBoost, 93.85% for Voting, and 99.38% for both MLP and ELM. Top- working models will help identify people in vulnerable positions toward addiction, evaluate the state of health and psychological condition of a person, which will be an indicator for acting in prevention. Outcome variables such as these from the current study are critical in deducing further risks for addiction, especially in targeted public health policies and intervention programs. With a special emphasis on current states of mental and physical health, the high performing models, especially MLP and ELM, proved as reliable tools for assessing risk for addiction. It outlines the role of machine learning in the transformation that views and methods toward the drug addiction crisis among Bangladeshi youth have been causing, with a data-driven basis for prevention and rehabilitation efforts. Project Mon, 13 Jan 2025 00:00:00 GMT http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/16815 2025-01-13T00:00:00Z