International Journal of Computational Thinking and Data Science

Exploring the Determinants of Facebook Ad Clicks among Malaysian Users using Machine Learning

2024-09-25T14:49:54+07:00

This research investigates the factors influencing Facebook users to click on a particular advertisement on the Facebook platform. This research has two main objectives. Firstly, it aims to analyze the relationships between Malaysian Facebook users’ advertisement click behavior with age, gender, day, frequency and engagement. Secondly, this research aims to compare the performance of various machine learning models in predicting Malaysian Facebook users’ advertisement click behaviour. The proposed strategy involves systematic data collection, data preprocessing, data analyzing, data splitting, data training, prediction and comparing performance of various machine learning models. By addressing these objectives, the research hopes to provide valuable insights for businesses to optimize their digital marketing. Engagement with the digital advertisement proves to be main factor in determining whether the user will click on the advertisement. Digital advertisements were also generally clicked towards the end of the week. Moreover, support vector machine proves to be the best model in predicting advertisement clicks among Malaysian Facebook users. In summary, the research was able to investigate the determinants of Facebook Ad clicks among Malaysian users using various machine learning models.

IoT-Based Monitoring and Optimization of Nutrient Film Technique Aquaponic Systems: Influence of Bio-Mechanical Filtration on Growth Rates of Oreochromis Niloticus and Brassica Rapa

2024-12-04T13:11:39+07:00

This study investigates the effects of bio-mechanical filtration on the growth of Oreochromis Niloticus and Brassica Rapa in a Nutrient Film Technique (NFT) aquaponic system, focusing on water quality and system monitoring. Aquaponics integrates fish farming and hydroponics into a symbiotic system that requires precise nutrient and waste management. Bio-mechanical filtration, essential for removing waste and maintaining water quality, was paired with an IoT-based monitoring system that tracked ammonia, nitrate, pH, and temperature in real-time. This enabled timely filtration adjustments to sustain optimal conditions for both fish and plants. Growth rates were analyzed across various filtration configurations, revealing that enhanced filtration and monitoring significantly reduced ammonia and nitrate levels, resulting in improved growth rates. These findings highlight the potential of combining bio-mechanical filtration and IoT monitoring to optimize aquaponic systems, making them more efficient, scalable, and resource-conscious. This study contributes to aquaponic design advancements, aligning with Sustainable Development Goal 2 (SDG 2) by promoting sustainable food production and food security.

YADIMS Enhanced Portal With AI Chatbot with Fine Tuning for Large Language Model

2024-11-12T13:47:27+07:00

The Yayasan Dakwah Islamiah Malaysia (YADIM) project, leveraging agile methodology, aims to develop an AI-powered chatbot and a digital portal to enhance digital interaction and educational outreach. Agile's iterative and incremental approach is crucial for overcoming the current limitations of YADIM's online platform, focusing on personalized and interactive user experiences through ongoing stakeholder collaboration and adaptability to evolving needs. The project's key phases, including requirement analysis, system design, and testing, will proceed in iterative cycles or sprints, each delivering a functional product increment. This method ensures continuous development and feedback integration, effectively serving diverse users like the public and staff with features like account management, content distribution, and efficient chatbot interactions. Agile principles guide the project's non-functional requirements, ensuring performance, usability, security, and maintainability with a focus on continuous improvement. The system architecture, following the Model-View-Controller (MVC) pattern, supports adaptability and scalability, aligning with agile's emphasis on technical excellence. A relational database evolves iteratively for efficient data management, and user-centric interface design is refined through regular user feedback. Concurrent development and testing guarantee the portal and chatbot's reliability and effectiveness. By adopting agile methodology, this project will significantly advance YADIM’s mission, spreading Islamic knowledge to a diverse Malaysian community with a flexible, responsive, and constantly improving digital solution. Future work aims to improve the system further. Potential enhancements include integration with payment gateways for online transactions, a customer feedback and review system, automated service reminders and notifications, and expanded service offerings and package management. These additions will enhance convenience, encourage a customer-centric service culture, ensure timely communication, and boost customer engagement and retention.

CyberShield Framework: Attacks and Defense Modelling for Cybersecurity in Internet of Things (IoT)

2024-12-23T17:25:18+07:00

The rapid emergence of Internet of Things (IoT) devices has ushered in an era of unprecedented connectivity and convenience, but it has also brought forth an array of cybersecurity challenges. This research deals with the multifaceted realm of cybersecurity threats on the Internet of Things (IoT) landscape. The vulnerabilities inherent in this ever-expanding ecosystem have become increasingly evident as IoT devices permeate every facet of our daily lives. In this paper, we review the various security threats that IoT deployments face, ranging from unauthorized access and data breaches to device vulnerabilities, physical attacks, and the pressing concerns surrounding user privacy. As the IoT continues to reshape our digital landscape, understanding and mitigating cybersecurity threats, in this paper we propose the attack and defense modelling framework to ensuring interconnected devices and systems' security, privacy, and resilience. This research serves as a foundational overview, urging stakeholders to prioritize IoT security in a world increasingly reliant on the promise of a hyper-connected future. By thoroughly synthesizing current knowledge in IoT, this research equips readers with a holistic understanding of IoT cybersecurity threats, enabling stakeholders to make informed decisions in a connected world. In the face of escalating risks, this research serves as a vital resource for academia, industry professionals, policymakers, and cybersecurity practitioners, all committed to fortifying the integrity and resilience of IoT ecosystems.

Harnessing AI for Comprehensive Retirement Planning and Preparation (AIRP2): A Multidimensional Approach to Financial, Career, Health, Social, Leisure and Spiritual Well-being

2024-12-23T15:16:15+07:00

This paper investigates the transformative potential of Artificial Intelligence (AI) in revolutionising retirement planning and preparation in Malaysia by integrating six critical dimensions: health, career, finance, interpersonal relationships, leisure, and spiritual well-being. The paper aims to examine the current state of retirement planning in Malaysia, identify prevailing gaps and challenges, and propose an AI-driven model for holistic retirement planning. This integrated approach addresses the multifaceted needs of employees, fostering a secure and fulfilling post-retirement life. Retirement planning is a pressing global concern, and Malaysia is no exception. As the population ages, there is a growing urgency for comprehensive retirement strategies that extend beyond conventional financial preparedness. Traditional methods often overlook essential aspects such as health management, career transitions, interpersonal relationships, leisure engagement, and spiritual well-being, leaving retirees vulnerable to multidimensional challenges. This conceptual paper has two primary objectives: (1) to analyse the multidimensional nature of retirement planning and its integration with AI; and (2) to develop a robust framework that incorporates AI technologies into retirement planning to enhance its overall effectiveness. The paper draws on theoretical insights and primary data, synthesising perspectives from peer-reviewed literature and foundational studies in the field. By bridging theoretical constructs with innovative AI applications, this research contributes to a deeper understanding of retirement planning, offering practical implications for addressing the evolving needs of Malaysia’s ageing workforce.