INTI Institutional Repository
Not a member yet
2178 research outputs found
Sort by
Lung Cancer Classification Using Stacking Framework of BiLSTM, Logistic Regression, and XGBoost
No full textLung cancer remains one of the most prevalent and deadly cancers worldwide, causing over 1.8 million deaths each year. Early and accurate classification of lung cancer is crucial, yet existing machine learning and deep learning models often face limitations in generalization and reliability. To address this issue, this study proposes a stacking framework that integrates Bidirectional Long Short-Term Memory (BiLSTM) and Logistic Regression as base learners, with Extreme Gradient Boosting (XGBoost) serving as the meta-learner. The rationale for this approach is that BiLSTM captures complex feature interactions, Logistic Regression provides interpretability, and XGBoost has demonstrated strong performance as a meta-learner in ensemble studies. The framework was evaluated on a publicly available lung cancer dataset consisting of 309 patient records with 15 clinical and lifestyle attributes. Experimental results showed that the stacking framework achieved perfect accuracy of 1.00, outperforming BiLSTM (0.95) and Logistic Regression (0.93). These findings confirm the effectiveness of the proposed ensemble in overcoming the weaknesses of individual models and highlight its novelty as a reliable approach for lung cancer classificatio
Unveiling Career Pathways: Success and Challenges of Bangladeshi Women in Computer Science Through Machine Learning
No full textAlthough the number of working computing women is steadily increasing in Bangladesh, it is a ray of hope that the gender gap is reducing day by day among final year students to higher-level job holders. This research aims to forecast how women in Bangladesh perceive and respond to pursuing careers in Computer Science. Primary data is collected by surveying women's experience that incorporates various open-ended and closed-ended questions and thus developed a dataset from 501 respondents, whereas respondents' age group were 19 to 60 years, and the majority were working in private sector jobs. A statistical tool Pearson's chi-square test is implemented to correlate between variables and thus different machine learning approaches, including Random Forest (which achieved a topmost accuracy of 85.00%), Decision Tree, XGBoost, Logistic Regression and K-Nearest Neighbors were implemented. It has explored the position, success, and obstacles of women in their place in Computer Science in Bangladesh, and one of the most delightful revelations is the borderline association of unequal pay. Notably, over 66% of the respondents reported that they do not encounter gender-based discrimination in their workplaces in terms of career advancement within various sectors of computer science
Technologization of Educational Standards and the Triple Hegemony of Cultural Inequality: Perspectives from Digital Education Internationalization
No full textThis study employs systematic literature analysis and multi-case comparison to investigate how the technologization of educational standards generates cultural inequality in the process of digital education internationalization. The analysis of 127 publications (2018–2023) yields three major findings: (1) Western-centric classification dominates nearly 90% of MOOC courses; (2) cultural conflicts account for a 31% dropout rate among Middle Eastern learners in virtual exchanges; and (3) China’s dual-narrative strategy significantly enhances cross-cultural acceptance (+41%). By introducing the Cultural Alienation Index (CAI), the study quantifies digital cultural inequality and validates its explanatory power through cross-regional cases. The findings contribute a triple-hegemony model (infrastructure, algorithms, knowledge output) and propose decolonization strategies, offering both theoretical insights and practical pathways for equitable digital education governance
A Literature Study on GenAI Adoption Behavior from the UTAUT Perspective
No full textBased on the Unified Theory of Acceptance and Use of Technology (UTAUT) and its extended frameworks, this study addresses the existing issues in current research on Generative Artificial Intelligence (GenAI) adoption behavior, including scattered research scenarios with insufficient cross-scenario comparisons, inadequate focus on dynamic adoption processes, gaps in research on specific population groups, and incomplete systematic application context of the UTAUT framework. Employing the literature review method, the study systematically retrieves and organizes literature related to the developmental history of the UTAUT theory and GenAI adoption behavior, aiming to clarify the research context of GenAI adoption behavior and identify the conclusions and limitations of existing studies. The research findings indicate that existing studies have confirmed that technological characteristics, individual traits, and environmental support are key factors influencing GenAI adoption, with widespread attention paid to trust and ethical issues. However, the aforementioned limitations still persist. Ultimately, at the theoretical level, the study proposes future research directions—strengthening cross-disciplinary integration, exploring dynamic evolution mechanisms, and expanding sample coverage—to enrich the relevant theoretical system. At the practical level, it provides targeted recommendations for enterprises, organizations, and policymakers to enhance the efficiency and rationality of GenAI adoptio
A Qualitative Study on AI Tool Adoption in Higher Education: A Cross-National Perspective
No full textArtificial Intelligence (AI) is currently integrated into most sectors, including education. AI tools within Higher Education (HE) have been shown to enhance students' academic performance, learning outcomes, and research productivity. The study addresses the adoption and application of AI tools in HE based on comparison of experiences of two nations. It is centered on exploring perceptions and challenges of teachers and students in integrating AI into pedagogical practices. Researcher adopting a qualitative research approach. Data were collected through interviews and focus groups such that there could be a deep understanding of the users' experience. Results show the cross-national difference in issues like awareness, accessibility, etc. for integrating AI. Research states that AI tools imply huge promise, and their successful use depends on digital literacy, policy contexts, and institutional readiness. On the basis of the result, the outcomes that contributed towards the policy makers are the Generative AI in the Academic syllabus at the tertiary level. In the future, a quantitative method will be added to this research to ensure high accuracy in this researc
Strengthening Ghana’s Banking Security through IoT: Implications for Achieving SDG Goals
No full textThe use of the Internet of Things (IoT) in Ghana's banking sector has enormously enhanced efficiency in operations and customer service delivery. The digitalization process, however, poses considerable cybersecurity threats. This study investigates the impact of IoT on Ghanaian banks' cybersecurity with emphasis on the prominent risks, challenges, and strategic interventions. According to secondary data extracted from top banking reports and qualitative analysis by IT security professionals, we have prevalence of threats such as distributed denial-of-service (DDoS) attacks, unauthorised access, and data breaches. The findings quote the highest priority need for robust cybersecurity measures, continuous risk assessment, and government-initiated regulatory structures to offset risks of IoT deployments in financial institutions. The findings reference the greatest need for efficient cybersecurity measures, continued risk evaluation, and regulatory-driven frameworks being initiated by governments in order to offset vulnerabilities of IoT installations in financial institutions It is in line with Sustainable Development Goal 9 (Industry, Innovation, and Infrastructure), SDG 11 (Sustainable Cities and Communities), and SDG 16 (Peace, Justice, and Strong Institutions), as regards providing safe, resilient, and secure financial environment
Investing in Sustainability: How FDI and Trade Openness Affect the Consumption of Renewable Energy in Regional Comprehensive Economic Partnership (RCEP) Countries
No full textThis paper focuses on how trade openness and foreign direct investment (FDI) affect the adoption of sustainable energy among the countries of the Regional Comprehensive Economic Partnership (RCEP) based on the panel data between 1998 and 2022. The Pesaran (IPS) and Phillips-Perron tests allow to assert the stationarity of data, and the Pedroni cointegration test indicates that there is a long-term correlation between the variables, which proves the theoretical connection between economic factors and renewable energy adoption. The research employs the Pooled Ordinary Least Squares (POLS), Fully Modified Ordinary Least Squares (FMOLS) and Dynamic Ordinary Least Squares (DOLS) to estimate long-run effects and Canonical Cointegration Regression (CCR) to assure robustness. The findings indicate that FDI has a very positive and significant impact on the usage of renewable energy, which indicates that FDI promotes the transition to cleaner energy in RCEP countries. Conversely, trade openness affects the utilization of renewable energy negatively, so that increased trade may be linked to the increased utilization of fossil fuels or the trade-offs related to the environment. The results offer useful policies that would be recommended to the RCEP countries, affirming the need to attract green FDI and use sustainable trade policies and reinforce the financial mechanisms that motivate the use of renewable energy in investments. This study is an addition to the wider debate of sustainable development and energy policy because it provides a great insight of the economic factors related to the use of renewable energy in developing nation
The Role of Learning Management Systems in Malaysian Universities: Advancing Quality Education and Reducing Inequalities
No full textThe successful use of Learning Management Systems (LMS) in higher education institutions is critical to improving the quality of teaching and encouraging student engagement. The paper presents trends, tendencies, and challenges of LMS use among Malaysian public and private university teachers. Employing a quantitative method, the research carried out surveys of teaching staff from different disciplines. The findings show stark contrasts between private and public institutions regarding LMS infrastructure, instructor preparedness, and pedagogical integration. By resolving these differences, LMS can provide an equitable digital learning environment in Malaysia by increasing access to high-quality education (SDG 4) and decreasing gaps between public and private schools or between urban and rural students (SDG 10). Private universities show high adoption rates, whereas public universities are held back by infrastructural and bureaucratic problems. The study provides recommendations to bridge the gaps and advance the digital teaching environment in Malaysian higher education
Deep Learning-Based Predictive Modeling for Male Depression Detection
No full textThis project utilizes machine learning techniques to construct a highly precise model for
categorizing audio recordings, with a particular focus on male speakers and their mental health
conditions. The audio recordings are classified into three distinct categories: Remitted (RMT),
Depressed (DPR), and High-risk for suicide (HRK), with special attention to gender-specific
nuances. We have conducted an extensive exploration and comparison of diverse machine learning
models, including 1D and 2D Convolutional Neural Networks (CNNs), Support Vector Machine
(SVM), and Recurrent Neural Network with Long Short-Term Memory (RNN-LSTM). Our
primary goal is to identify the most accurate model for classifying these male audio recordings,
potentially offering a valuable tool for early detection and intervention in male mental health
issues. We eagerly look forward to sharing our research results, aiming to make a substantial
contribution to the understanding and treatment of depression among males. In this paper, we
present the results of our investigation, comparing the accuracy of audio classification using 25-
second and 1-minute speech segmentatio
The Impact of Working Conditions on the Workers in Foundry Production
No full textThis article examines the impact of working conditions on the health of foundry personnel. The
study explores the effects of various production factors, such as dust, gas pollution, noise,
vibration, and microclimate parameters, on workers in different areas of foundry workshops.
The analysis also covers the state of occupational morbidity among personnel, taking into
account the type of production (mass, batch, and small-batch). The findings reveal that the
highest number of occupational diseases were detected in the cleaning, molding, melting and
casting departments of foundries. The most affected professions include casting dressers,
molders, smelters, melter-pourers, and repairmen. These results highlight the significant health
risks associated with working conditions in these areas and underscore the need for improved
safety measures and health monitoring in foundry environments