JOURNAL OF INFORMATION SCIENCE AND TECHNOLOGY

Prediction of PM2.5 Dust Concentration Using Machine Learning Techniques

Panida Mahahing, Pradthana Minsan — Sat, 28 Jun 2025 00:00:00 +0700

Fine particulate matter (PM2.5) is a significant air pollution issue, particularly in urban areas with high emissions, such as Bangkok, Thailand, and Guangzhou, China—one of the most populous cities in the world. This research focuses on developing a predictive system for PM2.5 concentration using machine learning techniques, including Linear Regression (LR), Support Vector Regression (SVR), and XGBoost, to aid in air pollution monitoring and management. The dataset used in this study is a secondary source containing recorded PM2.5 values from Guangzhou, China, between 2010 and 2015. Experimental comparisons of the three models reveal that XGBoost demonstrates the highest predictive performance across all timeframes. Specifically, for the 1-hour ahead prediction, the XGBoost model incorporating historical PM2.5 averages and seasonal data achieved an R² of 0.6728, MAE of 12.06, and RMSE of 17.87, outperforming both LR and SVR. Furthermore, the predictive performance of all models declined as the forecasting timeframe increased, but XGBoost consistently outperformed the other methods in every scenario. The inclusion of seasonal information and historical PM2.5 averages significantly enhanced the model’s ability to predict future PM2.5 concentrations.

Developing an Ontological Knowledge Base for Meaningful Search of Fish Transformation Based on Community Wisdom

Nattavut Sriwiboon, Achara Sumungkaset, Jetsada Singthongchai — Sat, 28 Jun 2025 00:00:00 +0700

This paper aims to design and develop a knowledge base and semantic search system for fish processing based on community wisdom. The study involves two groups of samples: (1) 21 individuals engaged in fish processing as either their primary or supplementary occupation, selected through voluntary sampling, and (2) 100 relevant documents on fish processing, chosen based on completeness of knowledge description to be stored in the ontology-based knowledge base. The search efficiency was evaluated using precision, recall, F-measure, mean reciprocal rank (MRR), and sum of squared errors (SSE). The findings indicate that the ontology-based knowledge base for fish processing consists of nine classes. The evaluation results show that assigning keyword weights using correlation coefficient analysis yields an MRR of 0.926 and an SSE of 0.091. Additionally, document similarity measurement for fish processing knowledge reveals that Method A achieves an F-measure of 0.8283, outperforming Method B, which has an F-measure of 0.6749. However, Method A has a lower mean precision of 0.7839 compared to Method B, which achieves 0.8324, indicating that Method B provides better search accuracy.

The development of Project Management System, Case Study: Software Engineering, Faculty of Liberal Arts and Sciences, Sisaket Rajabhat University

Supachai Thongsuk, Shatit Phetyoying, Chaiwat Boonon — Sat, 28 Jun 2025 00:00:00 +0700

Currently, information technology plays a crucial role in enhancing the efficiency of management processes, especially in student project management. Studies have shown that integrating information systems into project management can significantly streamline processes and allow students and faculty to conveniently track project progress. To improve the existing workflow efficiency, this research aims to: 1) examine communication processes and student project management patterns, 2) develop a web application for project management and tracking, and 3) study user satisfaction and test the system effectiveness. The research findings reveal that: 1) the web application segments operations according to user groups, including teacher and student, 2) the system effectiveness, as evaluated by system development experts, is rated highly on average, and 3) the overall user satisfaction with the web application among the sample group is also rated highly.

Development of a Line Chatbot Application for Promoting the Information and Communication Technology Program at Ubon Ratchathani University

Chanchai Supaartagorn, Fapraw Cheableam — Sat, 28 Jun 2025 00:00:00 +0700

This research aims to 1) design and develop a Line chatbot application for promoting the Information and Communication Technology program at Ubon Ratchathani University. 2) assess the users’ satisfaction with a Line chatbot application for promoting the Information and Communication Technology program at Ubon Ratchathani University. The researchers utilized Dialogflow, which incorporates natural language processing, and the LINE application for developing the chatbot. Additionally, Google Apps Script and Google Sheets were used for data management. The primary services of the LINE chatbot application include querying general information about the program and querying information about the courses in the program structure. System testing results indicated an overall user satisfaction rating of "very high," with an average of 4.52 and a standard deviation of 0.60. Therefore, this system will enhance the communication channels for those interested in the program and further promoting the program to a broader audience.

Development LINE Bot for Tourism Promotion in Tak Province

Sat, 28 Jun 2025 00:00:00 +0700

The research began with data collection on tourism in Tak and interviews with experts to gather requirements for the LINE bot’s functionality. Information from these expert interviews was used to define the functionality of the LINE bot. The design and development process involved creating graphics for display and menu interactions within the LINE application. The tourism data was designed and stored in MySQL. A web application was developed using HTML, CSS, JavaScript, and PHP to manage and present Tak’s tourist data via the LINE platform. Messaging API and LIFF were used for interaction between the user and the LINE bot. An evaluation of tourist satisfaction with the LINE bot was conducted, revealing a high level of user satisfaction.

Kids Learn to Love Tha Kham Community Culture Application to Enhance Learning Skills in the 21st Century in the Learning of Life and Professional Skills Through Community Culture

Khontaros Chaiyasut, Pimchanok Suwannasri, Praisun Suwannasr — Sat, 28 Jun 2025 00:00:00 +0700

This research aims to develop learning media and develop the Kids Learn to Love Tha Kham Community Culture Application. The research results found that the learning media obtained from the knowledge extraction from the community's wisdom consisted of 10 topics: collecting sweet vegetables with red ant eggs, making marigold flowers, making brooms, the tradition of carrying iron to support the bodhi tree, hand-woven karen fabric, natural colored woven cotton, local doctors, lanna language, murals, oil paintings, and traditional massage. The content was used to develop a hybrid application using the ionic framework. The results of the trial and quality assessment of the application found that the usability, feasibility, correctness, and appropriateness were at the highest level. It helps preserve the community's cultural wisdom, enhances learning skills, is free to download, and is consistent with real conditions. The content is developed from reliable theoretical concepts, is easy to use and install, and is easy to understand. The manual is appropriate. Therefore, the developed application is considered to be effective and acceptable.

Color Palette associated with Color Discrimination and Clustering for the Elderly and Normal Vision People

Thitiporn Lertrusdachakul, Suharit Vinijchayakul, Kasem Thiptarajan — Sat, 28 Jun 2025 00:00:00 +0700

Color is a powerful tool to communicate both direct and indirect information. It enhances readability, attracting attention and emotion. However, the use of color to deliver sophisticated information is still in challenge. This paper proposes an algorithm to generate color palettes to convey the message of group information comprising various levels of detail and interrelation. The algorithm is based on color data such as contrast ratio, lightness, chroma, hue, color distance. The results provide the color palettes associated with color discrimination and clustering for the elderly and normal vision people. This contributes to an improved visual experience in digital media, content, printing, and user interface, with particular relevance for the visual representation of group information, including status management and informational categorization.

Integration of Gamification in Virtual Reality Media to Enhance Solar System Learning

Puwadol Sirikongtham, Pramuk Boonsieng, Apichaya Nimkoompai — Sat, 28 Jun 2025 00:00:00 +0700

The objective of this study was to compare the effectiveness of a virtual-reality (VR) learning environment with conventional instruction and to develop a VR-based multimedia system on the solar system, in order to help students acquire knowledge and understanding of the solar system’s science. Students would learn the names and order of the planets, their sizes, their rotational and orbital velocities, and the number of moons orbiting each planet. The VR system was developed in C# according to the software development life cycle (SDLC). The system was evaluated by five experts and sixty students enrolled in MSC-112 Science and Technology, who were divided into two groups of thirty. The control group received traditional instruction, while the experimental group used the newly developed VR multimedia system. System quality and learning outcomes were analyzed using descriptive statistics—including percentages, means, and standard deviations. Results showed that the academic achievement in the solar system topic for the VR group achieved effectiveness scores of 82.43/81.23 (mean = 17.73, SD = 3.52), which was significantly higher than that of the traditional group (mean = 8.95, SD = 1.58) at the .05 level of statistical significance.