J-INTECH ( Journal of Information and Technology)

Application for Mental Health Consultation with Scheduling Function at the Counseling Guidance of Universitas Teknologi Yogyakarta

Kevina Maydiva Heriansaputri, Moh Ali Romli — Wed, 18 Jun 2025 07:29:45 +0000

This study developed a mobile-based mental health consultation application at Universitas Teknologi Yogyakarta (UTY) to offer both online and offline counseling services for students. The tool incorporates two primary features online consultations and offline consultation scheduling, designed to improve students' access to mental health support. The system utilizes Flutter for cross-platform mobile application development, Firebase for data management, and Node.js for backend services. The study employs a Research and Development (R&D) methodology encompassing needs analysis, system design, implementation, and testing. The results indicate that the application successfully mitigates obstacles such time limitations, stigma, and restricted accessibility, hence enhancing student involvement with mental health services. The Self-Reporting Questionnaire 20 (SRQ-20) serves as a mental health screening instrument within the application, enabling students to evaluate their mental health status. This program aims to deliver a thorough and accessible solution for mental health counseling at UTY and may serve as a prototype for other universities.

Application of the A-Star Method for Evacuation Routes Using the Long-Range Wide Area Network (LoRaWAN)

Charles Daniel, Moh Ali Romli — Wed, 18 Jun 2025 08:16:05 +0000

Optimal evacuation route planning is a crucial factor in disaster mitigation, especially in areas with limited communication infrastructure. This study proposes the application of the A-Star method for evacuation route optimization, supported by the Long-Range Wide Area Network (LoRaWAN) as an emergency communication system. The research methodology includes the development of an A-Star algorithm optimized with an adaptive heuristic function, integration with LoRaWAN-based sensors for real-time road condition monitoring, and simulations in various disaster scenarios. The results show that the developed system can reduce evacuation time by up to 31.4% compared to conventional methods and maintain communication connectivity up to 95% even under emergency conditions. Furthermore, the dynamic adaptation mechanism allows for automatic route changes based on current field conditions, enhancing the effectiveness of the evacuation process. Therefore, the integration of the A-Star method and LoRaWAN network proves to be a reliable and efficient solution for improving public safety during disasters.

Application of Honeypot in Network Security for Detecting Cyber Attacks on IT Infrastructure

Carlos Susanto, Moh Ali Romli — Wed, 18 Jun 2025 09:04:59 +0000

The high security risks that are susceptible to hacking and exploitation by malicious actors to steal data or information often arise due to a lack of awareness regarding the critical importance of implementing deceptive network security using honeypots. Negligence can create vulnerabilities that are easily exploited, allowing attackers to initiate breaches. A notable network security approach involves using Honeypots, a method that creates a decoy server to mimic an authentic one. Honeypots are deliberately engineered to attract the attention of cyber attackers and facilitate their access to the trap server, thereby enabling the monitoring and analysis of their activities without compromising the integrity of the primary server. To achieve optimal network security, comprehensive testing of Honeypots is essential. This testing process serves as a fundamental metric in evaluating the efficacy and performance of Honeypot systems in mitigating cyber threats.

Combination of Response to Criteria Weighting Method and Multi-Attribute Utility Theory in the Decision Support System for the Best Supplier Selection

Mon, 23 Jun 2025 07:33:41 +0000

Choosing the right supplier is a strategic factor in supporting operational efficiency and a company's competitive advantage. This process requires a decision support system that is able to assess various alternatives objectively and in a structured manner. This study aims to develop a decision support system in the selection of the best supplier by combining the Response to Criteria Weighting (RECA) and Multi-Attribute Utility Theory (MAUT) methods. The RECA method is used to objectively determine the weight of each criterion based on the variation of data between alternatives, so as to reduce subjectivity in the weighting process. Meanwhile, the MAUT method functions to calculate the total utility value of each supplier based on the normalization value and weight that has been obtained. The results of the RECA method show the objective weight of each criterion, which is then used in the MAUT calculation process. The results of the analysis, obtained in the best supplier selection based on the total score of each candidate, it can be seen that PT Global Niaga Mandiri ranks first with the highest score of 0.6512, this shows that this company is the best choice in the supplier selection process. In second place is UD Anugrah Bersama with a score of 0.399, followed by PT Indo Logistik Prima in third place with a score of 0.3451. The combination of the RECA and MAUT methods has been proven to be able to produce accurate, rational, and accountable decisions. This system provides a measurable approach in filtering supplier alternatives efficiently and is relevant to be applied to various other multi-criteria decision-making contexts.

Analysis of the Effectiveness of Traditional and Ensemble Machine Learning Models for Mushroom Classification

Neny Sulistianingsih, Galih Hendro Martono — Mon, 23 Jun 2025 08:08:37 +0000

The classification of edible versus poisonous mushrooms presents a critical challenge in the domains of applied biology and public health, particularly due to the serious implications of misidentification. This research employs the UCI Mushroom Dataset to evaluate and compare the effectiveness of several machine learning models, including traditional algorithms like Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, K-Nearest Neighbors and Naïve Bayes, as well as advanced ensemble techniques such as Stacking and Voting Classifier. Notably, both Random Forest and Stacking achieved flawless accuracy, reaching 100%, underscoring the high predictive capacity of these models in complex categorical scenarios. Conversely, Naïve Bayes exhibited significantly weaker performance—achieving only 59.8% accuracy—likely due to its underlying assumption of feature independence, which does not hold for this dataset. The ensemble learning approaches, including the combination of Stacking and Bagging, not only preserved but also enhanced model robustness and generalization. These methods effectively leverage the complementary strengths of individual learners to yield more accurate and stable predictions while mitigating overfitting risks. Comparative analysis with previous research confirms the consistency of these findings and reinforces the viability of ensemble strategies for handling intricate classification tasks. Overall, this study highlights the importance of algorithm selection tailored to data characteristics and supports the use of ensemble learning to boost predictive reliability.

Decision Support System for Selecting BPS Central Tapanuli Partners Using the SMART Method

Adzkia Nur, Ardilla Syahfitri Lubis, Dicky Sambora, Debi Yandra Niska — Mon, 23 Jun 2025 08:47:41 +0000

The selection of partners at the Central Statistics Agency (BPS) of Central Tapanuli is a very important process because it determines the quality of supporting staff in census and survey activities. One of the core stages in the selection process is the interview, which functions to directly evaluate the abilities and character of prospective partners. The assessment in the interview covers several main aspects, namely analytical skills, communication, appearance, and politeness. This study aims to design a decision support system based on the SMART (Simple Multi Attribute Rating Technique) method that can help process interview results systematically and objectively. Each criterion in the interview is given a weight based on the level of importance, then the value of each candidate is processed through mathematical calculations that produce a final score. This score is used to determine the candidate's ranking and provide recommendations to the selection committee. The system is developed in the form of a web-based application with a user-friendly interface, and supports data input, value processing, and automatic presentation of results. The implementation results show that the SMART method is able to improve assessment accuracy, reduce subjectivity, and accelerate the decision-making process in partner selection. With this system, the interview process is not only a fairer and more transparent means of assessment, but also supports work efficiency and consistency of selection results in the BPS environment.

Analyzing Students' Interest in Mathematics Through the Implementation of the K-Means Clustering Algorithm

Tue, 24 Jun 2025 09:45:37 +0000

This Research is motivated by the importance of understanding students' interest in mathematics, especially in State Junior High School 193 East Jakarta, considering that mathematics is often considered a difficult and frightening subject for some students. Learning interest, which is defined as the tendency of students to pay attention with a feeling of pleasure, has a significant influence on the process and results of student learning. This study aims to identify the level of student interest in mathematics using the K-Means algorithm. This method is used to group students into several clusters based on their level of interest. The results showed that students were divided into three clusters, namely the first cluster with very high interest totaling 193 students with an average Final Semester Exam score of 91.920, the second cluster with low interest totaling 18 students with an average score of 52.333, and the third cluster with high interest totaling 66 students with an average score of 87.606.

Designing a Web-Based Laundry Service Application For Strala Laundry

Cut Mutia, Fauziah Fauziah, Irzon Meiditra, Fitra Yuda, Rizky Rahmansyah — Sat, 28 Jun 2025 11:44:57 +0000

As daily activities become more hectic, the need for efficient laundry services is increasing. Currently, Starla Laundry relies on a manual system for operations, leading to challenges such as storing customer data in logbooks, creating complex reports, and slow transaction processes due to manual calculations. These issues make it difficult for the owner and admin to manage the business effectively amidst growing competition. To address this, a new application is proposed using PHP programming, a MySQL database, and Microsoft Visual Studio Code for data processing. The application aims to improve operational efficiency, simplify data management, and support business growth. It is expected to enhance customer satisfaction and open opportunities to optimize the business in the digital era. This study focuses on designing a laundry service information system using the System Development Life Cycle (SDLC) methodology, which includes planning, analysis, design, and implementation. The system was tested using Black Box Testing to ensure the software meets its requirements. The results confirm that all application functions work as intended, making it ready to streamline Starla Laundry’s operations efficiently and professionally.

Application Of K-Nearest Neighbor Algoritma for Customer Review Sentiment Analysis at Ngeboel Vapestore Shop

Muhammad Aryanda, Ita Arfyanti, Yulindawati Yulindawati — Sat, 28 Jun 2025 11:50:55 +0000

This study applies the K-Nearest Neighbor (K-NN) algorithm to classify customer sentiments from online reviews about Ngeboel Vapestore, a local MSME in the vape industry. A total of 175 reviews from Google Review and Instagram were processed using standard NLP techniques and TF-IDF for feature extraction. The best K-NN model (k=3) achieved 85.4% accuracy. Although Logistic Regression achieved higher accuracy (92.6%), it failed to detect negative sentiment. The findings highlight the potential and limitations of K-NN for sentiment analysis in underexplored MSME contexts like vape retail. The study recommends further model improvements and broader MSME applications.

Application of K-Nearest Neighbor Algorithm For Sentiment Analysis On Free Fire Online Game Based On Google Play Store Reviews

Dimas Raya, Amelia Yusnita, Ivan Haristyawan — Mon, 30 Jun 2025 02:32:41 +0000

The swift expansion of the digital gaming sector, especially online games like Free Fire, has produced extensive user feedback via platforms like the Google Play Store. This research utilizes the K-Nearest Neighbor (KNN) algorithm to conduct sentiment analysis on 5,000 user reviews, with the goal of assessing its classification effectiveness. Following preprocessing (case folding, Text Cleaning, tokenization, stopword Removal, stemming), the data was converted using TF-IDF and balanced through SMOTE. Experimental findings indicate that KNN attained a peak accuracy of merely 36.53% (at k = 14), reflecting weak performance with high-dimensional textual data. In contrast, Logistic Regression attained a notably higher accuracy of 88%, showcasing its dominance for this task. The results offer perspectives for game developers to assess user feelings and emphasize the significance of selecting suitable machine learning models. Future research should investigate advanced classifiers like SVM, Random Forest, or deep learning methods to enhance accuracy.

Development of a Web-Based Decision Support System for Selecting the Optimal Duck Feed Using the Analytic Hierarchy Process (AHP)

Khaira Nazla, Enrico Vincentius Manurung, M Fauzan Hidayat, Debi Yandra Niska — Mon, 30 Jun 2025 02:52:30 +0000

The selection of optimal livestock feed is essential for improving animal health and productivity. This study developed a web-based Decision Support System (DSS) using the Analytic Hierarchy Process (AHP) to help farmers choose the best feed based on nutritional content, price, and availability, including sub-criteria like protein, energy, fat, minerals, and vitamins. The system ranks alternatives using pairwise comparisons and priority weights. Validation against manual calculations showed high accuracy (correlation coefficient: 0.9987; errors <2.5%), with Fermented Feed (A3) as the top choice (score: 0.3611). Both methods produced identical rankings. The system reduces evaluation time by ~85% while maintaining accuracy, proving AHP’s effectiveness in digital livestock feed management tools.

Preventive Attendance Record using Photo from Mobile Phone and Printed Paper using CNN

Mon, 30 Jun 2025 00:00:00 +0000

Face-based attendance systems are increasingly popular for their ease of use, but they are susceptible to fraud, such as using photos or videos for unauthorized attendance. This study introduces a digital attendance system that combines facial recognition with liveness detection powered by Convolutional Neural Networks (CNN). Liveness verification is achieved by analyzing subtle movements and responses to ambient lighting. The dataset includes 30 facial images, encompassing both authentic and fraudulent samples. Testing demonstrates a facial recognition accuracy of 91.3% and effective spoofing detection in static and dynamic settings. This system provides a secure, fraud-resistant attendance solution ideal for educational and corporate settings. Further enhancements are suggested to improve performance across diverse facial expressions and lighting conditions.

Decision Support System for Determining Social Assistance Recipients in Petuaran Hilir Village Using the SMART Method

Debi Yandra Niska, Azura Calista Sitorus, Syafira Istiara, Rahma Hidayanti — Mon, 30 Jun 2025 03:49:50 +0000

The distribution of social assistance in rural areas is a strategic government effort to reduce social inequality and improve the welfare of underprivileged communities. However, in Petuaran Hilir Village, the process of determining aid recipients is still conducted manually, leading to various issues such as a lack of objectivity, potential unfairness, and mistargeting. Therefore, this study aims to design and implement a Decision Support System (DSS) using the Simple Multi-Attribute Rating Technique (SMART) method to determine social assistance recipients in a more systematic and transparent manner. The SMART method was chosen due to its effectiveness in simplifying multi-criteria decision-making and its practicality for implementation at the village level. The system was developed as a web-based application and tested using the black-box method, as well as validated against the manual selection results conducted by village officials. Testing results showed that the system can objectively identify and rank aid recipients based on final scores from five main criteria: income, number of dependents, home ownership status, housing condition, and type of employment. The system achieved 100% consistency with manual selection results and reduced the selection process time by up to 70%, enabling a fairer and more targeted distribution of aid based on systematically calculated scores. By eliminating manual bias in the selection process, the system significantly improves the accuracy of recipient rankings. This study also opens opportunities for further development, such as integrating real-time population data and advanced analytical features to support more responsive social policies.

Application of Naive Bayes Algorithm for Analysis of User Reviews on Mobile Legends Game: Bang Bang

Al-Muchlis Syachrul Ramadani Roba, Siti Lailiyah, Amelia Yusnita — Mon, 30 Jun 2025 06:05:24 +0000

Mobile Legends: Bang Bang is a highly popular MOBA game, especially among students, which generates a large volume of user reviews on the Google Play Store. These reviews provide a valuable data source for understanding user sentiment. This study conducts sentiment analysis on user reviews using three variants of the Naïve Bayes algorithm: BernoulliNB, GaussianNB, and MultinomialNB. From an initial 5,000 reviews collected via web scraping using Python, 4,428 reviews were used after neutral reviews were removed to focus solely on positive and negative sentiments.The preprocessing steps included case folding, word normalization, tokenization, stopword removal, and stemming. Sentiment labeling was carried out using a lexicon-based approach, comparing the frequency of positive and negative words in each review. The dataset was split in an 80:20 ratio for training and testing.The results show that MultinomialNB achieved the highest accuracy at 75%, followed by BernoulliNB with 74%, and GaussianNB with 50%. MultinomialNB demonstrated superior performance in detecting positive sentiments, while BernoulliNB offered more balanced results. GaussianNB performed poorly due to its assumption of normally distributed continuous data, which is unsuitable for text classification. This study concludes that Multinomial Naïve Bayes is the most effective model for sentiment analysis of user reviews when working with word frequency-based representations.

Development of Geolocation-Based Employee Attendance Application on Android Mobile

Nisa Kurnia, April Lia Hananto, Tukino Tukino, Shofa Shofiah Hilabi — Mon, 30 Jun 2025 06:37:55 +0000

The development of mobile-based systems in Indonesia has provided innovative solutions to improve the efficiency of conventional administrative processes, especially in employee attendance. This research aims to develop an Android-based employee attendance application that is integrated with geolocation technology to enable accurate and real-time attendance monitoring. This system is built using the Waterfall method, which includes the stages of needs analysis, system design, implementation using Flutter and Dart programming language, and testing using black box testing techniques. Black-box testing was conducted on six main functions, resulting in a 94% overall success rate. Most functions achieved a 100% pass rate, but two test cases for attendance check in/out failed due to GPS location inaccuracies, highlighting the impact of device and environmental factors. The average response time was 1.28 seconds, and the average GPS delay was 2.1 seconds. The implementation of real-time notifications and admin verification improved transparency and minimized attendance fraud. The results demonstrate that the application provides an effective and efficient solution for employee attendance management. Future work should focus on enhancing location accuracy, conducting non-functional testing, and expanding features to ensure broader adoption and system robustness.

Development of Academic Community Recommendation System Using Content-Based Filtering at UIN Malang Informatics Engineering Study Program

Mon, 30 Jun 2025 07:01:50 +0000

The mismatch between the number and quality of Information and Communication Technology (ICT) talents and industry needs in Indonesia creates significant challenges, especially for Informatics Engineering students who often experience difficulties in determining the appropriate professional field. This research aims to develop a content-based filtering-based academic community recommendation system to help students choose communities that are relevant to their interests, skills and experience. The system uses TF-IDF and cosine similarity methods to match student profiles with community descriptions. Data was collected from 48 students and 10 academic communities in the Informatics Engineering Study Program of UIN Malang, and processed through preprocessing stages before modeling. Evaluation results using the System Usability Scale (SUS) resulted in a score of 76, which is categorized in the “good” level, However, users indicated the need for improved guidance in navigating the system. This system is expected to be an innovative solution to increase student participation in appropriate academic communities, as well as support the development of their potential and readiness for the world of work

Post-Disaster Building Damage Segmentation Using Convolutional Neural Networks

Revaldi Rahmatmulya, Agung Teguh Wibowo Almais, Mokhamad Amin Hariyadi — Wed, 02 Jul 2025 00:00:00 +0000

Natural disasters are events caused by nature such as earthquakes, tornadoes, tsunamis, forest fires, and others. The impacts of natural disasters are significant and varied across various sectors, including the economy, health, and primarily, infrastructure. Effective and efficient actions are needed to assist in the recovery following natural disasters, one of which is aiding in the identification of building damage levels post-disaster. To address this issue, this research proposes a system capable of performing segmentation to determine the level of building damage post-natural disaster using convolutional neural network methods. The data utilized consists of aerial images sourced from xView2: Assess Building Damage, comprising 50 aerial images with 5 classes: no-damage, minor-damage, major-damage, destroyed, and unlabeled. The steps undertaken in this research include data preprocessing using patchify and data augmentation. Subsequently, feature extraction is performed using convolution, followed by the training process using a neural network with the proposed architecture. This study proposes an architecture with 27 hidden layers, with feature extraction utilizing average pooling. The model evaluation process will employ Mean Intersection over Union (MIoU) to assess how closely the segmentation prediction results resemble the original data. The proposed architecture demonstrates the best MIoU result with a value of 0.31 and an accuracy of 0.9577.

K-Means Clustering Algorithm Measuring the Satisfaction Level of MNC TV Muslim I'murojaah Program Viewers

Herayati Herayati, Andronias Siregar, Hariyanto Hariyanto — Mon, 07 Jul 2025 04:29:36 +0000

Television remains one of the most influential mass media platforms for disseminating information and entertainment to the public, with one notable example being the I’Murojaah program aired by Muslim TV on MNC Channels. The objectives of this study are to identify the level of viewer satisfaction with the I’Murojaah program, determine the indicators influencing viewer satisfaction, classify viewers based on their satisfaction levels, and provide recommendations to program managers to improve quality and viewer satisfaction. This study employs a qualitative approach using the K-Means Clustering algorithm. The data used in this study were obtained through a survey distributed to 100 respondents, covering several viewer satisfaction indicators such as content quality (6 questions) and program presentation (6 questions). The collected data were then grouped and analyzed. The results of the first cluster iteration distance calculation consisted of 84 viewers who were very satisfied, while the second cluster consisted of 16 viewers with relatively low satisfaction levels. The Davies Bouldin Index values were -0.674 for class 2 clustering, -2.001 for class 3 clustering, -1.961 for class 4 clustering, and clustering class 5 (-2.000). In conclusion, the best clustering performance results were for classes 2, 3, 4, and 5. The smallest Davies Bouldin Index value was for clustering class 2. Recommendations for program improvement include enhancing image and sound quality, ensuring that the equipment and technology used can produce optimal quality.

Quality Analysis of an Interactive Programming Learning Platform Based on ISO/IEC 25010 Using a String-Matching Approach on User Reviews

Mon, 07 Jul 2025 08:27:52 +0000

This study aims to analyze user perceptions of Quality in Use of the Khan Academy e-learning platform, focusing on two key characteristics defined in the ISO/IEC 25010 standard: satisfaction and efficiency. Based on the analysis of user review data, there is a clear difference in volume between the two aspects: 539 reviews (65%) reflected the satisfaction aspect, while only 290 reviews (35%) were related to efficiency. This indicates that users are approximately 1.86 times more likely to comment on satisfaction than on efficiency. For satisfaction, most of reviews were positive (420 reviews, or 77.9%), while 119 reviews (22.1%) expressed negative sentiments. These results suggest that most users are satisfied with their experience using Khan Academy, particularly due to factors such as flexible access time, user convenience, and the wide availability of learning materials. In contrast, the efficiency- related reviews exhibited a more even distribution, with 154 positive reviews (53.1%) and 136 negative reviews (46.9%). This closer balance indicates that while some users appreciate the platform's performance, others report encountering technical issues, including slow access speeds, navigation difficulties, and system instability. Overall, user perception of the Khan Academy e-learning system is generally positive, especially regarding satisfaction. However, the findings also underscore the importance of addressing technical performance challenges to improve efficiency and ensure a seamless learning experience. These insights provide a valuable basis for the development of user-centered e-learning systems and contribute to the evaluation of system quality from the Quality in Use perspective.

Developing Kindergarten Management System for Supporting the Regional Educational Report

Anita Anita, Sutan Arif Pambudi, Nira Radita, Hilman Nuril Hadi — Mon, 07 Jul 2025 09:07:32 +0000

Administrative tasks in early childhood education centers often face challenges when conducted manually, leading to delays in reporting, data inconsistencies, and limited access for parents. Unlike conventional school management systems, this study specifically targets kindergarten-level reporting that aligns with regional government formats (Korwilkersatdik) in Indonesia. To address these issues, this study focuses on designing and developing a web-based information system to help manage administrative and reporting activities at TK Dharma Wanita 91 Pesanggaran, a kindergarten in Banyuwangi, Indonesia. The research applied a descriptive qualitative method and followed a structured software development process, including system analysis, design, implementation, and evaluation. The system was built using PHP with the CodeIgniter framework and MySQL as the database. It includes modules for student registration, payment validation, attendance tracking, lesson planning, and monthly report generation. Each module was tested using black-box testing and successfully met functional requirements. The implementation of the system significantly reduced manual workload, improved reporting accuracy, and facilitated better communication with parents. The results indicate that the system effectively supports the institution’s operational needs and aligns with reporting standards set by local education authorities. Future development may focus on adding mobile accessibility and analytical features to provide broader access and deeper insights for school management. This study demonstrates how a domain-specific information system can significantly improve the efficiency of kindergarten administration while meeting government reporting standards.