https://ptsldigital.ukm.my/jspui/handle/123456789/388954 2026-05-09T04:56:15Z 2026-05-09T04:56:15Z Bashar Hamad Aubaidan (P103708) https://ptsldigital.ukm.my/jspui/handle/123456789/783262 2026-05-07T07:23:32Z 2026-02-24T00:00:00Z

Title: Improving the quality of dataset for diabetes prediction using a machine learning approach Authors: Bashar Hamad Aubaidan (P103708) Abstract: Diabetes is a chronic metabolic condition marked by persistently elevated blood glucose levels and influenced by multiple physiological and lifestyle factors, making early prediction challenging. Reliable predictive modelling requires high-quality datasets; however, issues such as missing values, class imbalance, and redundant or high- dimensional attributes often reduce model accuracy and generalisability. To address these limitations, this study introduces an integrated data-quality enhancement framework combining Bidirectional Neighbour Graph (BNG) imputation for missing data, Clustering Selection Synthesis Filtering (CSSF) for class imbalance, and Rough Set Theory (RST) for dimensionality reduction and feature selection. The proposed BNG–CSSF–RST framework was applied to three independent datasets: the Pima Indians Diabetes Dataset from the UCI Repository, the Diabetes Clinical Dataset (Kaggle, 2024), and the Kaggle Diabetes Prediction Dataset (2023). After preprocessing with the proposed framework, both Artificial Neural Network (ANN) and Support Vector Machine (SVM) models were trained and evaluated. Across all datasets, the framework yielded notable improvements in predictive performance. Using the Pima dataset (768 records, 9 features), ANN achieved 93.51% accuracy, while SVM reached 90.26%. For the Diabetes Clinical Dataset (100,000 records, 17 features), ANN obtained 96.95% accuracy and SVM achieved 96.32%. On the Kaggle Diabetes Prediction Dataset (100,000 records, 9 features), ANN attained 91.49% accuracy, and SVM achieved 89.12%. Overall, the results indicate that systematically addressing missing data, class imbalance, and irrelevant or redundant features substantially improves classification performance. The enhanced accuracies observed across all three datasets exceed those typically reported in earlier studies, confirming the robustness of the proposed BNG–CSSF–RST framework. Finally, this approach provided a methodology for diabetes prediction specifically designed to address missing data, class imbalance, and dimensionality reduction, thereby enhancing overall data quality and enabling more robust analysis of complex datasets.

2026-02-24T00:00:00Z Ahadi Haji Mohd Nasir (P127430) https://ptsldigital.ukm.my/jspui/handle/123456789/783261 2026-05-07T07:04:34Z 2025-12-30T00:00:00Z

Title: An ontologically founded and governance-integrated information dashboard design method Authors: Ahadi Haji Mohd Nasir (P127430) Abstract: Information dashboards have become essential monitoring and decision-support tools in smart waste management systems, providing stakeholders with real-time operational insights to enhance efficiency and effectiveness. Despite their widespread adoption, existing dashboard design practices remain fragmented and ad hoc, with most studies focusing on domain-specific implementations using diverse approaches. Critically, no standardized Information Dashboard Design Method (IDDM) exists that integrates Information Governance (IG) principles or employs ontological approaches, resulting in inconsistent design practices, limited knowledge reuse, and insufficient attention to governance requirements such as data quality, compliance, and accountability. This research addresses this gap by developing the IDDM Canvas, an ontology-based method that embeds IG principles to enable a systematic, accountable, and reusable dashboard design approach that reduces reliance on ad hoc practices. The research employed Design Science Research Methodology (DSRM) and synthesized common elements from thirteen recent dashboard design studies into twelve core building blocks organized across three categories: Defining and Planning, Design and Visualization, and Testing and Improvement. These building blocks are governed by eight IG principles: Effectiveness, Transparency, Accountability, Quality, Integrity, Compliance, Security, and Availability, which are integrated within the Information Dashboard Design Ontology (IDDO). The IDDO was developed using the Unified Ontology Approach (UOA), a customized Ontology Development Methodology (ODM), and formalized using OntoUML grounded in the Unified Foundational Ontology (UFO). The IDDM Canvas operationalizes IDDO into a structured, governance-aware method demonstrated through the Intelligent Monitoring Recycle System (i-MORSYS), where two independent designers applied the IDDM Canvas to create operational and analytical dashboards on Tableau and Looker Studio. Evaluation through semi-structured interviews with the two designers and one waste management expert, analyzed using Braun and Clarke's thematic analysis, revealed positive outcomes across seven themes: Framework usability and learning experience, Enhanced design thinking and user-centricity, Comprehensive coverage with implementation gaps, Practical implementation challenges, Framework flexibility and adaptability, Recommendations for enhancement, and Overall assessment and adoption potential. Participants rated the framework's usefulness at 8.5/10, with all expressing willingness to adopt the method in future projects. This research contributes theoretically through the invention of IDDO and the improvement of dashboard design method through the IDDM Canvas, positioning both within the Design Science Research contribution matrix. Practically, the framework provides a reusable, standardized method bridging ad hoc design practices and governance-aware, systematic approaches. The integration of IG principles establishes a foundation for ensuring data quality, regulatory compliance, and organizational accountability. The framework's ontological foundation enables potential adaptation beyond waste management to other domains requiring governance-aware dashboard design. Future work includes developing iterative design support, operationalizing governance principles with actionable guidance, and exploring the IDDM Canvas's potential as input for AI-assisted dashboard generation tools.

2025-12-30T00:00:00Z Muhammad Yudhi Rezaldi (P81308) https://ptsldigital.ukm.my/jspui/handle/123456789/781714 2025-12-09T08:05:27Z 2020-10-12T00:00:00Z

Title: Aesthetics 3D geovisualization model for flood disaster using overlap XYZ coordinate Authors: Muhammad Yudhi Rezaldi (P81308) Abstract: Floods are a common type of natural disaster, which are as a result of global warming and are amongst the most harmful to people and property. Prior awareness and information supply to the community is required to reduce the negative impacts of floods. One way to overcome this problem is by providing a community with education media that can help in delivering the disaster information. 2D and 3D geovisualization techniques have been used in flood modeling to convey flood disaster information. However, visual results stemming from such techniques are still unsatisfactory. This study has applied the aesthetic values of the elements to produce 3D geovisualizations which seem more realistic. 3D geovisualization software such as ArcGIS is capable of producing format files that are compatible with multimedia software, which can be used to add aesthetic value to the visualizations. In the study, the geovisualization process was performed by transforming spatial data into visuals using ArcMap. The results of the visuals were then combined with other spatial data using ArcScene. To make the visuals of 3D geovisualization appear more realistic, 3D object mapping methods were implemented using photogrammetry techniques that were produced from aerial photos using drones. The 3D geovisualization was combined by an overlapping method of XYZ coordinates with Cinema 4D software. The modeling results of overlapping XYZ coordinates were further refined by re-modeling and adding aesthetic parameters using several multimedia software. In conducting the experiment of developing the prototype in this study, data from the flood that had occurred in Jambi city in year 2013 was used as the sample data. The prototype was used to evaluate the proposed Aesthetic 3D Geovisualization Model to introduce the aesthetic value in building the realistic 3D geovisualization visuals. In the stage of evaluation, a questionnaire was developed and a survey using it was conducted, to evaluate the success of the Aesthetics 3D Geovisualization Model in delivering the information of flood disaster for a community education program. A total of 100 respondents comprising of visual communication designers were involved in the questionnaire, where they were required to compare three geovisualized 3D samples using non-probability sampling. The evaluation results showed that 84.6% of the respondents chose Aesthetic 3D Geovisualization to produce the flood models which could be used to convey flood disaster information. The reliability statistics table from validation shows that Cronbach's Alpha value is 0.935, which is significant enough to validate that the usage of aesthetic parameters in disaster information media for flood modeling is able to make the visualization of flood modeling more realistic and preferable by respondents. Moreover, the information is more communicative in conveying visual cues and is also easily understood and accepted by the audience. Therefore, the Aesthetic 3D Geovisualization Model is highly recommended to be used to convey information about floods. The contribution of this study is the production of visualizations based on spatial and non-spatial data that have applied aesthetic values by an overlapping method of XYZ coordinates, thereby creating a new concept of multimedia design through storyboards, by adding some aesthetic value. The method of overlapping XYZ coordinates is new for geology, limnology, and for finding formulations for creating flood height animations based on data in the multimedia field. Description: Full-text

2020-10-12T00:00:00Z Yasuki Shima (P81310) https://ptsldigital.ukm.my/jspui/handle/123456789/781712 2025-12-09T08:02:25Z 2020-01-30T00:00:00Z

Title: Optimization of public transport bus scheduling using kmeans and genetic algorithm Authors: Yasuki Shima (P81310) Abstract: Along with economic growth in the developing countries, the population of urban areas is increasing tremendously. Alongside this, the public transport bus service is expected to develop with the growth of urbanization. For that reason, optimization of the scheduling of public transport such as public buses is an important task in the operation of public transport. In scheduling public transport operations, it is difficult to plan for optimal service provision that can apply to different travel areas, time zones, cycle frequency, and scheduling of vehicles and crew. Basically, public transport design consists of four plans: network design, timetabling, vehicle scheduling, and crew scheduling. The timetabling is important in those plans for financial reasons as well as meeting passenger demand for a reliable service. Several methods have been proposed to deal with the task of public transport scheduling. For example, some research has been conducted to shorten the waiting time of passengers and to review cycles in areas where there are few passengers, but these still could not meet the passenger demand. Therefore, providing the optimum scheduling of public transport, designed accurately according to the demand, is a means to shorten the waiting time of passengers and eliminate useless bus cycles. This research started with investigating and clustering the time zones by using K-means based on the collected GPS data of public buses. The GPS data was processed to produce three attributes known as “Time”, “Volume” and “Quality”. Then, a genetic algorithm was implemented to optimize the public transport scheduling. The result of the experiment shows that the proposed optimization method increases the frequency of bus cycles during peak passenger hours in some time zones and conversely reduces the frequency of bus cycles for low passenger hours. The proposed method is able to both optimize the financial benefit and meet passenger demand. The result of the experiments shows that the proposed method is better than existing methods, increasing by about 30% the optimization accuracy. The dataset for the experiments used was GPS Public Bus in Okinawa, Japan. Description: Full-text

2020-01-30T00:00:00Z