Journal of Applied Informatics and Technology

Journal of Applied Informatics and Technology (Editorial Note)

Journal of Applied Informatics and Technology — 2025-12-07

The Journal of Applied Informatics and Technology (JIT) has published Volume 8, Number 1 (2026): January-June, comprising a total of 13 research articles. Each article has undergone peer review by at least three experts from different institutions. The authors revised their articles based on the reviewers’ comments until they were approved for publication. The editorial team also verified the accuracy of the content, references, and language usage to ensure the highest quality. We sincerely thank all the reviewers for their valuable feedback and thoughtful recommendations.

Adaptive Routing using Relative Cost Concept in Elastic Optical Networks with Space-Division Multiplexing

Anwer Alyatama — 2025-10-19

Elastic Optical Networks with Space Division Multiplexing Networks (EON-SDM) is the most prominent technology for efficiently transmitting massive amounts of data flow and dealing with the explosion of new technologies that require a wide range of services. Among the various factors that determine network efficiency, performance, and design effectiveness in EON-SDM networks is routing, spectrum, and core allocation (RSCA). RSCA's strategic optimization ensures efficient data transmission, effortlessly meeting the demands of high-bandwidth applications. Within this study, we propose a relative cost-based adaptive RSCA algorithm that prioritize long-term state consequences in accepting new calls. When a call is received, the proposed approach picks the route, core, and spectrum group with the lowest relative cost and compares it to the call’s value, in the event that the minimum computed relative cost of the arrival call is greater than the call value, the call is blocked. Comparative analysis with established routing methods reveals promising performance outcomes, including a 45% reduction in normalized loss revenue and a 155% enhancement in fairness in certain scenarios.

The Development of an Automatic Environment Control System in Greenhouses for Phoenix Oyster Mushrooms using the Internet of Things

Pensiri Phoriya — 2025-10-19

Mushrooms are rich in protein, nutritional substances, and health-enhancing medicinal substances, making them a popular choice for consumption. This popularity has spurred a growing interest in mushroom cultivation. However, mushrooms thrive only within specific temperature and humidity ranges. Deviations from these conditions can adversely affect their growth and overall production. Therefore, this research aimed to develop an automatic environmental control system in greenhouses for Phoenix oyster mushrooms using the Internet of Things and evaluated the system's performance by experts while also assessing user satisfaction. The system is designed to control temperature and humidity to be at the optimum level for mushroom growth. It sends alerts to users if the temperature exceeds a specified threshold. The system utilizes NodeMCU to connect with temperature and humidity sensors and controls equipment such as water sprayers, ventilation fans, heat bulbs, and lights inside the greenhouse. Accessible via a web application, the system employs software development tools, including Arduino IDE, Firebase real-time database, LINE Notify, and Google Sheets. Statistical analysis relied on mean and standard deviation. The results showed that the system could effectively control greenhouse conditions to cultivate Phoenix oyster mushrooms through web applications and send LINE notifications. This may help reduce damage to Phoenix oyster mushroom cultivation that could occur. Evaluation results from experts scored were at the highest level, with an average of 4.79, while user satisfaction evaluations were also at the highest level, with an average score of 4.80.

Detecting Fish Movement Behavior based on Constant Temperature and pH using the IoT and Realtime Object Detection Model

Hariphat Phongsuwan — 2025-10-19

This paper introduces a novel approach to water quality monitoring utilizing fish behavior as an indicator of aquatic environmental conditions. The proposed system integrates advanced technologies, including computer vision, sensors, and data analytics, to track and analyze the movement patterns of fish in real-time. Deployed sensors measure key water quality parameters such as temperature, pH, dissolved oxygen, and turbidity, while cameras capture fish activities within the monitored water bodies, then the livelihood of the fish will be affected. With problems that arise, solutions can be prepared by analyzing key components. Two components consist of a system that can automatically set fish feeding and the second component is water property measurement and level monitoring using the sensor module and electrical equipment with automatic video recording equipment installed, in observing the behavior and livelihood of aquatic animals. It has a communication connection in the form of the Internet of Things (IoT). We have developed a system that allows devices to measure temperature and pH in water and to display via the Internet for use in online data tracking and automatic installation of video recording equipment. By correlating observed fish behaviors with concurrently recorded water quality data, the system establishes meaningful relationships that enable the continuous assessment of the aquatic environment. From the results, it was known that the best conditions for the fish to survive. The project can be used as a prototype for creating an automatic fish tank that can adjust various parameters within the water if it is not appropriate for the well-being of aquatic animals. This innovative approach contributes to the advancement of environmental monitoring systems, fostering a deeper understanding of the dynamic interplay between aquatic ecosystems and water quality parameters.

Transfer Learning-Based Segmentation of Pneumonia from Chest X-Rays Images

Kyi Pyar — 2025-10-19

Pneumonia remains a significant global health concern, warranting precise and efficient diagnostic tools. This study introduces a comprehensive approach to pneumonia segmentation leveraging advanced deep learning techniques. The primary goal is to enhance the precision of pneumonia localization within medical images, specifically chest X-rays, through the utilization of state-of-the-art deep learning models. This study explores the application of advanced segmentation models, namely DeepLabv3 and SegNet, for the automated identification and delineation of pneumonia-affected regions within chest X-ray images. DeepLabv3, renowned for its semantic segmentation capabilities, and SegNet, featuring an encoder-decoder architecture, are selected as the segmentation models. The training process of the system leverages the widely acknowledged Kermany dataset, specifically composed of chest X-ray images depicting cases of pneumonia. This dataset is well-established and holds a prominent status within the field, recognized for its relevance and significance in the context of pneumonia detection and classification tasks. As per the evaluation findings, it is evident that the system attains enhanced accuracy when employing the DeepLabv3 architecture compared to the SegNet architecture.

Boosting Multi-Object Tracking Performance with Advanced Attention Mechanism based on Transformer

Pimpa Cheewaprakobkit — 2025-10-19

The Transformer architecture has been successful for a variety of tasks in natural language processing and has recently gained popularity in computer vision applications such as medical image analysis, traffic light monitoring, surveillance, and object tracking. This is due to the self-attention operation in Transformer, which enables global interactions between image patches, resulting in powerful modeling capabilities. However, the quadratic complexity in time and memory associated with the Transformer poses a challenge. To address this issue, we propose an advanced attention mechanism based on the Transformer architecture for multi-object tracking. The advanced attention mechanism consists of Transposed Self-Attention (TSA) and Cross Patch Interaction (CPI) modules. The TSA module utilizes a transposed self-attention mechanism instead of the conventional self-attention, operating across feature channels rather than image patches. It has a linear computational complexity with respect to the number of patches, resulting in reduced computational costs. Meanwhile, the CPI module is added following the TSA module to enable communication across patches, thereby enhancing the model's learning efficiency. This approach can efficiently process high-resolution images, achieving state-of-the-art performance in multi-object tracking with a MOTA of 72.8% on MOT17.

Autoencoder Applications for Enhancing Spice Classification Performance under Limited Dataset Conditions

Evan Tanuwijaya — 2025-10-19

Indonesia's rich assortment of spices holds immense value for their health benefits and culinary applications, yet widespread understanding of these spices among the populace remains limited. This research proposes leveraging autoencoder technology to address the complexities arising from the intricate shapes and colors of spices, compounded by the constraints of a limited dataset. Through meticulous refinement of hyperparameters such as learning rate, optimization function, input size, and epochs, the autoencoder endeavors to extract unique features crucial for accurate spice classification. Comparative assessments against alternative CNN architectures employing transfer learning underscore the autoencoder's capacity to attain comparable accuracy levels. Notably, the model demonstrates consistent performance devoid of overfitting or underfitting, yielding validation precision and recall rates nearing 0.97. These outcomes accentuate the autoencoder's efficacy in compressing image data and discerning pivotal features, thereby facilitating precise spice classification despite constraints posed by dataset limitations.

Adaptive Deep Neural Network for Solving Multiclass Problems

Komgrich Onprasonk — 2025-10-19

In recent years, multiclass classification has gained significant attention due to its wide-ranging applications in fields such as healthcare, finance, and image recognition. The ability to accurately classify data into multiple categories is essential for developing intelligent and robust systems. This research compares the performance of several machine learning and deep learning algorithms for multiclass classification tasks, with a focus on adaptive techniques in neural networks. The evaluated algorithms include Support Vector Machines (SVM), One-vs-Rest Logistic Regression (OvR-LR), Deep Neural Networks (DNN), Dropout-enhanced DNN, and Adaptive Regularization-based DNN. The experimental evaluation was conducted using both the train–test split and 5-fold cross-validation methods to ensure result reliability and generalizability. The Adaptive Regularization-DNN model achieved the highest performance among all tested approaches, with an accuracy of 98.75% under the train–test split and 97.3% under cross-validation. These results highlight the model’s robustness and its effectiveness in minimizing overfitting in structured multiclass classification problems. Performance metrics including precision, recall, F1-score, and accuracy were used to provide a comprehensive evaluation of each model’s capabilities.

Advancing Dermatological Care through AI: A Deep Learning-Based LINE Chatbot for Skin Disease Diagnosis

Prem Enkvetchakul — 2025-10-19

This paper presents the development and deployment of a LINE chatbot for diagnosing skin diseases using advanced deep learning techniques, addressing the challenge of timely and accurate diagnosis in resource-limited settings. While previous studies have explored convolutional neural networks (CNNs) for medical image classification, our research distinguishes itself by integrating MobileNetV2, DenseNet201, and EfficientNetB4 architectures within a mobile messaging platform. These CNN architectures are well-recognized for their robustness in image analysis, yet few studies have harnessed their potential in real-time diagnostic tools accessible to the general public via widely-used platforms like LINE. Our chatbot facilitates the immediate assessment of user-uploaded images, offering a proactive tool for managing skin health. By leveraging deep learning, we enhance diagnostic accuracy and reduce the reliance on traditional medical consultations. Experimental results indicate that our approach not only improves diagnosis precision but also extends the accessibility of dermatological care, particularly in underserved regions. This work contributes to the growing body of mobile health technologies by showcasing the transformative potential of AI-driven solutions in healthcare. The novelty of our research lies in the seamless integration of cutting-edge CNNs into an easy-to-use chatbot, enabling real-time, remote diagnostics in a practical, scalable manner.

Impact of Urbanization on Urban Heat Island and Urban Thermal Field Variance Index of Coastal City Surabaya, Indonesia between 2005 and 2020

Risma Salsabila Foresty — 2025-10-22

The city of Surabaya serves as the capital of East Java Province, Indonesia, experiencing the most intensive anthropogenic activities in this province. It has emerged as a primary destination for urbanization from various regions, leading to notable shifts in land use, vegetation, and Land Surface Temperature (LST). This study aims to analyzing the impact of urbanization on urban heat island and urban thermal field variance of coastal city Surabaya, Indonesia between 2005 and 2020. The findings reveal that Land Use Land Cover (LULC) changes significantly impact LST in Surabaya. Pearson Correlation results showed a positive correlation between Normalized Difference Vegetation Index, Urban Thermal Field Variance Index, and Urban Heat Island in 2005, 2011, 2015 and 2020 (p < 0.01). Alterations in land use contribute to elevated surface temperatures owing to diminished vegetation and increased built-up areas, thereby exacerbating ecological degradation and posing health risks to the urban environment. Consequently, implementing mitigation and adaptation measures becomes imperative to address the escalating urban heat phenomenon. Strategies such as augmenting green infrastructure and optimizing blue infrastructure are crucial for maintaining thermal equilibrium across the urban landscape.

Enhancing the Performance of Sentiment Analysis Models Using GridSearchCV: A Case Study on Electric Vehicles in Thailand

Chirawat Emphan — 2025-10-25

This study investigates the enhancement of sentiment analysis model performance through hyperparameter tuning using GridSearchCV, with a focus on electric vehicle reviews in Thailand. To address the challenges inherent in Thai language processing, PyThaiNLP was employed for word segmentation and text preprocessing. Four machine learning models-Support Vector Machines (SVM), Multinomial Naive Bayes, Logistic Regression, and Stochastic Gradient Descent-were implemented for sentiment classification. The hyperparameters of each model were systematically optimized to determine the configuration that maximizes accuracy, precision, recall, and F1-score. Experimental results revealed notable improvements across all models following optimization. The SVM model, identified as the best-performing classifier, achieved an accuracy increase from 75.00% to 76.69% and an F1-score improvement from 74.82% to 76.69%. The optimal SVM configuration employed a radial basis function kernel with a regularization parameter (C) of 10 and a gamma value of 0.1. These findings underscore the significance of hyperparameter optimization in improving model effectiveness and contribute to advancing sentiment analysis in linguistically complex environments such as Thai.

The Development of Brand Equity Measurements for Competitiveness of Thai Seafood Industry in Consumers-Perspective

Nutthapon Jitprapai — 2025-10-26

The study aimed to develop a precise measurement of brand equity for the Thai seafood industry, specifically focusing on canned tuna and shrimp products, from a consumer perspective. It utilized a comprehensive mixed-methods approach that combined qualitative interviews with experts and a quantitative survey of 500 consumers, following Churchill's recommendations. The structural validity of the research variable measurement was tested. Research Findings: The study revealed a consumer-oriented brand measurement model comprising four key components: brand loyalty, brand awareness, brand perceived quality, and brand association. This model provides valuable insights for assessing and generalizing brand equity within the Thai canned tuna and shrimp market. Theoretical Contribution/Originality: This research contributes significantly by offering a reliable measurement of brand equity for canned tuna and shrimp brands in Thailand. The comprehensive model, encompassing brand loyalty, awareness, perceived quality, and association, holds practical implications for marketers and practitioners aiming to enhance brand equity within the industry.

The Semantic Web of Digital Technology and Learning in the 21st Century for Undergraduate using Ontology Techniques

Sawitree Pipitgool — 2025-10-29

In the 21st century, digital technology has become integral to daily life, significantly impacting the skills and knowledge of undergraduate students. This research aims to develop a Semantic Web for learning digital technology in the 21st century by employing ontology techniques to enhance the efficiency of information retrieval. The system is designed to offer flexible learning, adaptable to students' needs, and focuses on categorizing content into three main classes and twelve subclasses. These classes define relationships using four object properties to connect main classes, subclasses, and instances, and four data type properties to link instances with data and relationships between digital technologies. This approach clarifies information and makes it more relevant for undergraduate students. Despite the advantages of ontology techniques in improving information retrieval and recommendation processes, challenges remain due to the complexity of constructing data relationships and establishing rules for data storage and retrieval. Effectively managing semantic data requires specialized knowledge to ensure accurate and efficient outcomes. The ontology knowledge base primarily consists of digital technology, innovation, and digital skills. Based on evaluations by three experts, the Semantic Web for digital technology learning in the 21st century, developed using ontology techniques, was rated at a very good level (\bar{x} = 4.52, S.D. = 0.19). The system's performance was also validated, showing precision at 96.25%, recall at 92.08%, and an F-measure of 95.29%, indicating its effectiveness in supporting learning through digital technology.

Tamarind (Tamarindus indica L.) Charcoal Color for the Use in Woodcut Printmaking

Woraunyu Narongdecha — 2025-10-24

This research aimed to study the quality of color produced from tamarind (Tamarindus indica L.) charcoal for use in woodcut printmaking and to compare the color fastness of woodcut prints on various types of paper by examining four color formulas blended with mixtures of Arabic gum and tamarind seed gum. The results revealed that Fabriano paper was the most suitable material for woodcut printmaking using tamarind charcoal color, particularly with Formula 1, which consisted of tamarind charcoal powder, Arabic gum, glycerin (C3H8O3), honey, and deionized water (H2O), as well as Formula 2, which additionally contained indigo (Indigofera tinctoria L.) powder. The comparison of hue and fastness, as determined by a colorimeter, showed that Formula 2 on Fabriano paper exhibited the highest L* value and intensity at 20.80. The +a value indicated a shift toward the green zone, while the –b value indicated a shift toward the blue zone, reaching its highest level at –1.90 due to the addition of indigo powder. Moreover, the average intensity (L* value) on Fabriano paper across the four color formulas showed consistent results. Applying these colors in the woodcut printmaking process demonstrated smooth hues with high intensity, durability, and detailed color expression. In contrast, the color blended with tamarind seed gum showed unstable endurance when applied with a paint roller, resulting in unclear details and hues due to color coagulation. All four color formulas demonstrated good solubility and could be used with either water or deionized water as solvents.