INTI Institutional Repository
Not a member yet
2178 research outputs found
Sort by
Humanizing AI in Higher Education: Rethinking HR, Ethics, and Skill Development for University Employees in the SDG Era
No full textThe rapid integration of Artificial Intelligence (AI) in higher education presents a profound dilemma of how universities can balance technological innovation with ethical governance and sustainable human resource (HR) practices. This study addresses the ethical considerations in AI adoption, the lack of accountability and transparency in AI-assisted HR functions, and the skills gaps among university employees transitioning to an AI-supportive environment. Equity, bias, and preparedness are central to institutional integrity and employee wellness in digitally transforming traditional higher education organizations. This qualitative study employed semi-structured interviews with academic heads, human resource managers, and administrators, supported by a review of institutional policies, accreditation models, and AI governance reports. It sought to understand how universities are redefining their HR, ethics, and capacity building processes to integrate AI humanely in academic and operational spaces. The results identified three predominant themes: (1) unbalanced ethical frameworks and low transparency in AI-driven hiring and performance appraisal; (2) faculty and staff apprehension due to algorithmic biases, data privacy issues, and a lack of institutional oversight; and (3) an urgent need to develop emotional intelligence, interdisciplinary thinking, and digital sensitivity among employees. Based on these findings, the paper proposes a Humanizing AI in Higher Education Framework focusing on ethical accountability, diversified HR policies, and continuous professional learning, aligned with the United Nations Sustainable Development Goals (SDG 4 - Quality Education, SDG 8 - Decent Work, and SDG 16 - Peace, Justice, and Strong Institutions). This model offers strategies for universities to balance technological advancement with human values in the AI revolutio
Comparative Study on Water Potability Prediction using Ensembled Based Techniques
No full textWater quality assessment plays a vital role in public health protection and environmental sustainability. Conventional testing techniques, though accurate, are time-consuming, labour-intensive, and prone to human error. Recent advancements in Machine Learning (ML), Deep Learning (DL), and the Internet of Things (IoT) have transformed water potability prediction through intelligent, automated systems. This paper presents a comparative review of ensemble and hybrid ML/DL approaches such as Bagging, Gradient Boosting, XGBoost, and stacked models that have achieved accuracies ranging from 83% to 99.6% in recent studies between 2023 to 2025. Furthermore, IoT-based sensors and blockchain integration enable real-time monitoring, transparency, and data security in water management frameworks. This work highlights current trends, research gaps, and emerging innovations focusing on adaptive, scalable, and secure water quality prediction systems for sustainable smart water management
The Challenges of Waste Management in Tourist Destinations: A Sociological Study
No full textThis study examines the growing challenge of waste generation in tourist destinations, focusing on Udupi district, known for its diverse landscapes and high tourist inflow. Tourism significantly increases waste especially, plastics and food waste leading to pollution, health risks, pressure on local waste systems, and conflicts between tourists and residents. Using a descriptive research design and simple random sampling of 150 respondents, the study found widespread negligence among tourists regarding waste disposal, even among the educated. All surveyed tourist spots face major recycling and reuse challenges. Despite efforts by the Department of Municipal and Panchayat Administration, diverse cultural behaviors, business activities, and inadequate amenities contribute to rising waste levels. The research highlights how convenience-seeking and limited awareness of local waste norms worsen the problem. It recommends promoting sustainable practices, strengthening waste infrastructure, and encouraging responsible tourist behavior to reduce environmental impacts and support effective waste management
Geomorphometric Insights into the Dharla and Teesta River Basins through GIS and RS Methods
No full textThis study employs morphometric analysis using Geographic Information System (GIS) and Remote Sensing (RS) techniques to evaluate and compare the geomorphometric and hydrological characteristics of the Dharla and Teesta river basins in northern Bangladesh. Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (30 m) data were used to delineate drainage networks and compute morphometric parameters from linear, areal, and relief aspects. Findings indicate that the Dharla River Basin (623.25 km²) is a fifth-order elongated basin characterized by moderate dissection, active erosion, and high flash flood vulnerability. In contrast, the Teesta River Basin (1,901.33 km²) exhibits a more complex drainage structure and regional-scale flooding patterns due to its Himalayan-fed origin and braided channels. By integrating SDG 6 (Clean Water and Sanitation) and SDG 13 (Climate Action), this research underscores the importance of morphometric assessment for sustainable watershed management, erosion mitigation, and climate-resilient flood control in northern Bangladesh
Design, Simulation, and Implementation of an IIoT-based Temperature and Humidity Monitoring System with Single-Core Infinite Loop Prevention and Fault Tolerance for Multi-Sensor and Connectivity Failures
No full textThe purpose of this research is to develop and implement a robust monitoring system that can function with the constraints of a single-core processing system, such as an ESP 8266, to overcome the issue of infinite loop faced when one or more subsystems malfunction, such as the loss of data or server connectivity. This paper describes a novel approach wherein the implementation of an Industrial Internet of Things (IIoT) based temperature and humidity monitoring system resolves the infinite loop issue and enables the development of highly robust monitoring solutions. The output from such a system can be subsequently integrated into other systems that can respond to the monitored values in real time without disruptions. The proposed mechanism utilizes an ESP8266 microcontroller for processing and wireless connectivity, DHT22 sensors for temperature and humidity measurements, an LCD for real-time monitoring, and the Message Queuing Telemetry Transport (MQTT) protocol to store data in the Adafruit IO platform for live off-site monitoring and data storage. Testing demonstrated that the system could handle up to three sensor failures out of four, Wi-Fi network and server disconnections, and automatic reconnections after five seconds. The system was able to perform error handling while maintaining the data flow from the sensors to the local data display LCD without interruptions during all tested scenarios. Based on several trials, the system succeeds at addressing a wide range of errors and disruptions, resulting in an ideal solution for the sectors that require precise monitoring to attain a wide range of operational objective
AI-Enabled Mobile Application for Surgical Safety Checklist Automation and Wrong-Patient Surgery Prevention
No full textWrong Site, Wrong Procedure and Wrong Patient adverse events (WSPEs) rank among the most alarming and catastrophic errors in surgical practice. Although such incidents are relatively rare, their impact becomes significant in highly populated countries like India, where the sheer volume of daily procedures amplifies the risk. The wrong procedure and wrong patient adverse event anomaly occur when the surgery is being done on the wrong person, often due to name similarities, leading to unwanted surgery that could cause a lot of damage. Although standardized surgical checklists are implemented to prevent such errors, high surgical volumes often make strict compliance challenging for healthcare teams. The increased workload can lead to oversight, reducing the effectiveness of manual verification processes. To address this issue, this paper presents an AI-driven mobile application employing FaceNet architecture to streamline checklist completion and enhance patient identification accuracy. By automating critical verification steps, the solution minimizes human error and reinforces adherence to safety protocols, ensuring more reliable and efficient surgical procedures. Experimental validation proves that the FaceNet achieved 100% detection and 98.33% recognition accuracy. Hospitals can integrate this technology to strengthen patient safety measures and reduce the incidence of preventable adverse events
Implementation of Artificial Mangrove Roots FADS in Preventing Abrasion
No full textFADS (Flotilla Abrasion Defense System) is coastal protection system designed to imitate mangrove roots, which serve to protect beaches and help retain sediment. As we know, it takes a long time for natural mangrove roots to grow strong and large. An innovation has been developed to create artificial mangrove roots, providing immediate access to the protective benefits of mangrove roots. FADS are made from environmentally friendly materials that are durable for up to 12 years. Testing and implementation have been carried out on sandy and muddy beaches in Thailand, and the results show that FADS perform effectively
Real Time Crowd Counting System Using Machine Learning
No full textCrowd counting is a critical task in public safety, event management, and urban planning. This paper presents a real-time crowd counting system leveraging machine learning to accurately estimate the number of people in a given scene. The proposed system employs a convolutional neural network (CNN)-based deep learning model, optimized for processing images and video streams to identify and count individuals in diverse environments. Key features of the system include real-time inference, robust performance in varying lighting and density conditions, and adaptability to different camera perspectives. The model is trained on a diverse dataset, encompassing crowded events, open spaces, and public gatherings, ensuring its versatility and reliability. Post-training, the system is deployed using lightweight architectures, allowing seamless integration with edge devices and IoT platforms
Gamification and Reward Systems for Enhancing Student Involvement in Extracurricular Activities at Universities
No full textGamification is a recent trend that seeks to raise people's motivation, involvement, engagement, and loyalty. It began as a marketing tool but has now expanded to several fields where public participation is crucial. Inspired by the widespread usage of video games, gamification seeks to increase game-like engagement by leveraging the game’s qualities in non-gaming situations. This study aims to highlight the research and knowledge regarding the use of gamification as a strategy to promote and enhance student engagement in higher education for extracurricular activities. The method used for this paper reviews general literature, which focuses on gamification and reward systems for enhancing student involvement in extracurricular activities. The conceptual findings suggest there is tremendous potential in gamification as a reward system for enhancing student involvement in extracurricular activities
Data Science for AI Chatbot Bias Detection and Mitigation in Healthcare
No full textThe integration of AI chatbots into healthcare systems presents transformative potential to enhance patient access, assist clinical decision-making, and streamline administrative workflows. Despite these advantages, the deployment of AI chatbots introduces significant concerns related to bias, which can diminish care quality and reinforce existing health disparities. This paper investigates the key sources of bias in AI chatbots, including dataset imbalances, algorithmic design flaws, and linguistic biases that may perpetuate stereotypes. These forms of bias can lead to misdiagnoses, inequitable treatment suggestions, and a breakdown of trust in AI-driven tools, particularly affecting marginalized or underserved populations. The study underscores the broader consequences of biased AI systems in healthcare, such as reinforcing discrimination and widening healthcare inequalities. To confront these challenges, the paper outlines methodologies for bias detection, including the use of fairness metrics and testing across diverse demographic cohorts. It also discusses mitigation strategies like representative data sampling, algorithmic refinement, feedback loops, and human oversight to ensure ethical and equitable AI usage