Journal of Science and Technology Buriram Rajabhat University (Online)

Development of Interactive Learning Materials Application in the Metaverse to Promote the Learning Potential of Information Technology for Life Courses Phetchabun Rajabhat University

วรรณภัสร์ ปราบพาลา — 2025-05-25

The objectives of this research were 1) to create and determine the effectiveness of interactive learning materials in the metaverse to promote learning potential, 2) to compare learning achievement before and after classes with interactive learning materials in the metaverse, and 3) to evaluate satisfaction with the use of interactive learning materials in the metaverse to promote learning potential. The sample was used for specific selection among students. The tools used in the research included 1) interactive learning media applications in the form of the metaverse, 2) academic achievement tests, and 3) satisfaction assessments for the use of learning materials. The data analysis used basic statistics such as mean (X), standard deviation (S.D.), analysis of the effectiveness of learning materials with E1/E2 criteria, and hypothesis testing with t-test.
The results showed that: 1) the developed interactive learning materials in the metaverse were as effective as 81.00/82.50, which met the set criteria, 2) learners had statistically significantly higher academic achievement after learning than before learning at the level of .05, and 3) Students' satisfaction with the use of interactive learning materials in the form of the metaverse was very high ( X = 4.33 , S.D. = 0.61)

Classifying Thai Social Media Opinions in The Covid Vaccine Using Text Mining

peakon maisook — 2025-05-25

This research aims to compare the efficiency of five modeling techniques: Naive Bayes, Support Vector Machine, Ripper, Fourier, and Random Forest. These techniques are used to build models for classifying the opinions of Thai people on social media regarding COVID-19 vaccination for children. The data, consisting of 2,466 opinions, was collected from online social networks. Verbs, adverbs, and adjectives were selected for model building, as these types of words can clearly indicate positive and negative sentiment. In this research, 10-fold cross-validation was used to divide the data into training and testing sets. Furthermore, accuracy, recall, and precision were calculated to compare the performance of the models. The experimental results show that Naive Bayes is the most efficient modeling technique, achieving an accuracy of 95.40%, a recall of 95.40%, and a precision of 95.40%.

Impact of rainfall on the infectious incidents in Kakasin province

Chatsiri Chatphuti — 2025-05-25

This correlational research aimed to analyze the impact between average monthly daily rainfall and the number of patients with rain-associated diseases in Kalasin Province. Rainfall data were collected from the National Hydroinformatics Data Center, Hydro-Informatics Institute (Public Organization) and monthly patient data collected from the Disease Surveillance Reporting System 506, Bureau of Epidemiology, Department of Disease Control. This comprised monthly data from January 2018 to July 2025. Data were analyzed using descriptive statistics, Pearson correlation, and generalized linear models. The analysis results revealed that the correlation coefficients between average monthly daily rainfall and the monthly number of patients with mushroom poisoning, hand-foot-mouth disease, and leptospirosis were 0.485, 0.346, and 0.233, respectively, indicating moderate, weak, and weak positive correlations. The generalized linear models with average monthly daily rainfall as the independent variable that best fit the data for the monthly number of mushroom poisoning patients and hand-foot-mouth disease patients were the negative binomial model and the quasi-Poisson model, respectively, with coefficients of determination (R²) of 0.338 and 0.193, respectively. Specifically, for mushroom poisoning, a 1-millimeter increase in average monthly rainfall was associated with a 1% increase in the number of patients.

Effects of seed pelleting with plant hormone GA 3 on quality of red cos lettuce seed

Purachai Sangkhao — 2025-05-25

ABSTRACT

This study investigated the effects of pelleting red cos lettuce seeds with the plant hormone Gibberellic acid (GA₃) on germination and seedling growth. The experiment was conducted using a completely randomized design (CRD) with 6 treatments and 4 replications each, at the Seed Technology Laboratory and the experimental greenhouse, Faculty of Agriculture and Agro-Industry. Talcum and Carbonate were used as pelleting materials, and Polyvinylpyrrolidone K90 served as a binder, along with different concentrations of GA₃. The results showed that pelleting seeds with GA₃ at a concentration of 0.02% resulted in the highest germination percentage and germination speed under greenhouse conditions. Furthermore, pelleting seeds at concentrations of 0.04% and 0.02% led to the highest average shoot length and root length compared to non-pelleted seeds, with highly significant statistical differences (P < 0.01). However, no statistically significant differences (P > 0.05) were observed in germination percentage and germination speed under laboratory conditions.

Bioactivity and Chemical Compositions of Salacca wallichiana Mart Leaf Oil

Salwa Torpee — 2025-05-25

This research aimed to investigate the biological activity and chemical composition of oil extracted from Salacca leaves using steam distillation. The chemical composition analysis using Gas Chromatography-Mass Spectrometry (GC-MS), revealed that the major component of the Salacca leaf oil was palmitic acid, accounting for 12.76% of the total composition. Additionally, other minor components, including various fatty acids and plant sterols were identified. The antimicrobial activity of the Salacca leaf oil was tested against pathogenic and multidrug-resistant microbial strains using the disc diffusion assay. The results showed that the oil exhibited antibacterial activity against all tested strains, including those producing Extended-Spectrum Beta-Lactamase (ESBL) strains. The Salacca leaf oil showed the most potent inhibition against Escherichia coli A1 (ESBL), with an inhibition zone of 10.83±0.56 millimeters. This study indicates that Salacca leaf oil is a local plant-derived substance rich in nutrients and possesses significant biological activity. It has potential applications in food and medicine.

Stepper Motor Damage Prediction Using Machine Learning Algorithms

noppharit sriwichai — 2025-05-25

This study presents a method for predicting stepper motor failures in modern automation systems by collecting data from five types of sensors: current, voltage, torque, temperature, and vibration, along with timestamp records and motion error measurements over a three-month period. The study compares the performance of three machine learning algorithms: Gradient Boosted Trees, Deep Learning, and Extreme Gradient Boosting. The results indicate that Gradient Boosted Trees achieves the highest accuracy at 91.17% and can predict a 90% probability of failure within 5 to 6 months. Feature importance analysis reveals that temporal factors and vibration have the most significant impact on motor degradation, accounting for 55.32% and 28.35%, respectively. These findings can be applied to predictive maintenance planning to minimize unplanned production line downtime and enhance overall production efficiency.

Keywords: Stepper Motor, Failure Prediction, Machine Learning, Gradient Boosted Trees, Predictive Maintenance

Supplier Selection with a 2-Tuple Linguistic Decision-Making Model: A Case Study of Vitamin Supplement Manufacturer

Pravee Sriprom — 2025-05-25

This research aims to select raw material suppliers in the vitamin supplement manufacturing industry using a two-tuple decision-making model. The decision criteria considered include price, quality, transportation, service, warranty, and environmental factors. Decision-makers utilized two-tuple linguistic variables to assess the importance of decision criteria in order to assign weight to each criterion. Subsequently, raw material suppliers were compared under various decision criteria to calculate two-tuple linguistic scores. The results revealed that Supplier A₂ had the highest suitability with a two-tuple linguistic score of (S₄, 0.23), followed by Supplier A₃ with a score of (S₃, 0.17), Supplier A₁ with a score of (S₃, -0.32), and Supplier A₄ with a score of (S₂, 0.37), which was the least suitable option. The outcome of this raw material supplier selection process was able to reduce non-compliant raw materials by 3,600 kilograms, with a total value of 3,477,600 THB.

Development of an ABICS System for Automatic Colony Counting of Escherichia coli and Enterobacter aerogenes from Smartphone Photographs

Phongsatorn Taithong — 2025-05-25

Counting colonies of Escherichia coli ATCC25922 (ECA) and Enterobacter aerogenes DMST2720 (EAD) is a crucial step in assessing the quality of food or raw milk. Manual counting typically takes approximately 2 - 5 minutes per Petri dish, depending on the colony density. This study presents an "Android Bacteria Image Counting System" (ABICS) that employs Projection Profile, Circle Hough Transform, and Power Law Transformation image processing techniques to enhance image clarity and accurately count colonies. In experiments using 84 Petri dish images, ABICS demonstrated an average counting accuracy of 90.77% when compared to manual counting, which generally exhibits an error rate of 5 - 10%. Significantly, ABICS required only 3 - 5 seconds per dish for colony counting, which is at least 24 times faster (averaging 35 - 100 times faster) than manual counting. Furthermore, ABICS significantly reduces the analysis time, making it a potentially valuable tool for enhancing efficiency and reducing workload in microbiological analysis.