Journal of Computer Networks, Architecture and High Performance Computing
Not a member yet
473 research outputs found
Sort by
Development of Disaster Management Information System Application with Five Integrated Features
Get PDFIndonesia, as one of the world’s largest archipelagic countries, is highly prone to natural disasters due to its geological, geographical, and socio-demographic conditions. Despite various efforts from the government and related institutions, there remains a lack of an integrated technological solution that addresses the critical aspects of disaster management. This research aims to develop a mobile-based disaster management information system called Aksi Bencana Indonesia, featuring five integrated functions: real-time disaster information, donation facilitation, volunteer management, logistics coordination, and disaster education. The development process involved identifying functional and non-functional requirements through interviews with disaster mitigation experts and literature studies, followed by UI/UX design using Figma, and coding using Flutter for the frontend and Laravel for the backend. User Acceptance Testing (UAT) was conducted with participants aged 18–55 to evaluate the system’s usability and effectiveness. The results showed that the application met the users’ expectations in providing timely and reliable information, facilitating donations and volunteer coordination, and supporting educational initiatives. Moreover, the integration of real-time features improved response time and enhanced the efficiency of resource distribution during disaster events. This study concludes that the application successfully bridges the gap between communities, donors, and volunteers, offering a practical solution for disaster preparedness and response. Future research may focus on expanding the system’s interoperability with national disaster databases and enhancing the analytics dashboard for better decision-making
Comparative Analysis of YOLOv11 with Previous YOLO in the Detection of Human Bone Fractures
Get PDFAccurate and rapid detection of bone fractures is an important challenge in the medical world, particularly in the field of radiology. This study aims to analyze and compare the performance of the YOLOv11 model with several previous versions of YOLO, namely YOLOv5, YOLOv8, and YOLOv10 in the task of detecting human bone fractures on X-ray and MRI images. The dataset used is the Human Bone Fractures Multi-modal Image Dataset (HBFMID) which consists of 641 raw images (510 X-rays and 131 MRIs). The four models were trained using the HBFMID dataset that had gone through a manual augmentation and annotation process, then tested using evaluation metrics such as precision, recall, mAP50, and mAP50-95. The training results showed that YOLOv11 has the most stable and consistent loss curve, with a fast convergence process. In terms of evaluation, YOLOv11 recorded a precision of 99.87%, a recall of 100%, a 99.49% mAP50, and an 84.13% increase in the number of mAP-95s, which generally outperformed other models. In addition, the visual prediction results show that YOLOv11 can detect fracture areas with the right bounding box and a balanced confidence score, without showing symptoms of overconfidence or inconsistency. When compared to approaches from previous studies, YOLOv11 also showed a significant improvement in detection accuracy. Thus, YOLOv11 is rated as the most optimal and reliable model in deep learning-based automatic bone fracture detection. This model has great potential to be applied in medical diagnosis support systems to improve the efficiency and accuracy of digital fracture identification
Usability Evaluation of the Online Marriage Registration Feature in SIMKAH
Get PDFThe development of digital technology has encouraged government agencies to adopt online-based public services. One such system is SIMKAH (Marriage Management Information System), a web-based platform initiated by the Indonesian Ministry of Religious Affairs to simplify marriage registration processes at Kantor Urusan Agama (KUA). However, several usability challenges remain, particularly in its online registration feature. This study aims to evaluate the usability of SIMKAH using the Think Aloud method based on ISO 9241-11, focusing on effectiveness, efficiency, and satisfaction aspects. The study involved 10 participants, all prospective brides and grooms who had never used SIMKAH before. They were asked to complete 10 scenario-based tasks while verbalizing their thoughts. The results show a high effectiveness rate of 87%, indicating that users were generally able to complete tasks successfully. However, efficiency was affected by lengthy form fields and confusing file upload sections. Satisfaction received a score of 4.1 out of 5, reflecting a positive experience overall, although users noted a lack of clear guidance, feedback notifications, and mobile optimization. User feedback highlighted the need for interface improvements, such as simplifying form structures, adding real-time validation, implementing autosave features, and enhancing visual guidance. These findings suggest that while SIMKAH is functionally adequate, improvements in usability are crucial to ensure a more seamless and satisfying user experience in digital public service
Integrated Cnn Based Facial Emotion Detection And Camera Based PPG Heart Rate Monitoring
Get PDFHuman emotion detection and heart rate estimation are two important aspects in developing a more responsive and adaptive human-computer interaction system. This study proposes a real-time video-based system that is able to detect facial emotions and estimate the user's heart rate simultaneously. The Convolutional Neural Network (CNN) method is used to classify facial expressions into several emotion categories such as happy, sad, angry, afraid, and neutral. Meanwhile, heart rate estimation is carried out using a non-contact Photoplethysmography (PPG) approach, which utilizes variations in color intensity in the user's facial area from video recordings to calculate the pulse rate. This system is developed using a standard webcam camera without additional medical devices, allowing for practical and economical implementation. The test results show that the system is able to recognize facial expressions with good accuracy, and estimate heart rate with an average error rate that is still within the tolerance limit of non-medical applications. By integrating computer vision technology and biometric signals, this study contributes to the development of a passive, real-time, and easily accessible emotion and health monitoring system
APPLICATION OF TRANSFORMER MODEL AND WORD EMBEDDING IN SENTIMENT ANALISYS OF INDONESIAN E-COMMERCE APPLICATION REVIEW
Get PDFThe rapid growth of e-commerce applications in Indonesia has resulted in a large volume of user reviews. The review contains important information that can be used to understand user satisfaction, complaints, and needs. Therefore, sentiment analysis of e-commerce app reviews is important to support future decision-making. This study aims to explore and compare the performance of the Transformer model and various word embedding methods in analyzing the sentiment of reviews of Indonesian e-commerce applications. The methods used involve extracting review data from the Google Play Store, text preprocessing, and text representation using Word2Vec, FastText, and IndoBERT. Next, this combination of embedding was tested using the Gradient Boosting Classifier as a prediction model. The evaluation was carried out by comparing the accuracy, precision, recall, F1-score, as well as the visualization of the confusion matrix and word cloud for each model. The results of the tests that have been carried out show that all three models have a fairly good ability to recognize positive reviews, with the highest accuracy score of 88% achieved by Word2Vec and FastText. While IndoBERT produces a lower accuracy value of 86%, IndoBERT shows a better balance in recall values and f1-scores for minority classes compared to Word2Vec and FastText. In conclusion, the application of the IndoBERT-based Transformer model is more effective in capturing the context and meaning of sentiment in Indonesian-language e-commerce reviews. These findings are expected to be a reference for the development of a more accurate sentiment analysis system for e-commerce applications in Indonesia.
 
Breast Cancer Classification Using Naïve Bayes and Random Forest Algorithms
Get PDFBreast cancer is one of the leading causes of death among women in Indonesia. Therefore, early detection is crucial to improving the chances of successful treatment. This study was conducted to evaluate the performance differences between the Naïve Bayes and Random Forest algorithms in classifying breast cancer data. The dataset used was sourced from Kaggle, and the entire data processing and model analysis process was performed using RapidMiner software. Data was split into 80% for training and 20% for testing to ensure optimal model evaluation. Evaluation was conducted using accuracy, precision, and recall metrics. The findings of this study indicate that Random Forest is capable of producing more effective classification performance than Naïve Bayes. Random Forest achieved an accuracy of 99.27%, recall of 99.27%, and precision of 99.30%. Meanwhile, the Naïve Bayes algorithm only achieved an accuracy of 83.78% with recall and precision of 83.80% each. The superiority of Random Forest is believed to stem from its ensemble approach, which can handle data complexity and reduce the risk of overfitting, thereby providing more accurate and stable prediction results. Based on these results, Random Forest is considered more suitable for use in machine learning-based early breast cancer detection systems. This study is expected to serve as a reference for the development of medical decision support systems and to encourage the use of classification technology in the field of health
TOURIST VISIT PATTERN ANALYSIS AT HOTELS IN NORTH PENAJAM PASER REGENCY USING K-MEANS CLUSTERING
Get PDFPenajam Paser Utara Regency, as a strategic area in East Kalimantan, has experienced significant development in the tourism sector in line with the plan to relocate the national capital (IKN). However, the utilization of tourist visitation data in hotels in this region is still not optimal. This study aims to analyze tourist visit patterns at Penajam Paser Utara Regency hotels using data mining techniques with the K-Means Clustering algorithm. The data used is secondary data obtained from the Penajam Paser Utara Regency Culture and Tourism Office, covering 34 hotels with variables including domestic and foreign visitors from 2019 to 2024. The clustering results show two main clusters: a high-visitation cluster comprising large hotels and a low-visitation cluster consisting of hotels with fewer visitors. The analysis reveals the dominance of domestic tourists, accounting for 99% of total visits, and the tourism sector's recovery pattern, reflecting a V-shaped recovery post-pandemic. This research contributes to hotel managers in designing market segment-based marketing strategies and local governments in designing data-driven tourism policies to enhance the sustainable competitiveness of destinations
Comparative Analysis of the Fuzzy Time Series Chen and Rungge Kutta Felhberg Methods for Forecasting the Number of HIV/AIDS in the Province of North Sumatra
No full textThe increase in HIV/AIDS cases in North Sumatra Province requires accurate forecasting methods to support prevention and control programs. Accurate predictions of the number of cases will help stakeholders design more targeted interventions and allocate resources effectively. This study aims to compare the performance of the Chen Fuzzy Time Series method and the Runge-Kutta Fehlberg numerical method in forecasting the number of HIV/AIDS cases in North Sumatra Province. The data used are monthly HIV/AIDS case data obtained from the North Sumatra Provincial Health Office. The Chen Fuzzy Time Series method is applied to capture patterns in data that are uncertain and ambiguous, while the RKF method is used to solve the logistic growth model that represents the development of cases. Forecasting accuracy was evaluated using the Mean Absolute Percentage Error (MAPE) metric. The results showed that the RKF method produced a lower MAPE value compared to the Chen Fuzzy Time Series method, indicating higher prediction accuracy. The RKF method provides more stable predictions for the next three months and is closer to the actual trend, while the Chen Fuzzy Time Series method shows slightly larger deviations but remains useful for imprecise data. In conclusion, both methods can be used for HIV/AIDS case forecasting, but the RKF method is proven to be superior in accuracy for the data used in this study
Evaluation of the Use of Web-Based Library Management System at I Gusti Bagus Sugriwa State Hindu University Denpasar
Get PDFLibraries as the heart of higher education play a vital role in supporting the Tri Dharma, particularly in academic and research development. The advancement of information technology demands innovation in library services, one of which is the adoption of Web-Based Library Management Systems (WLMS). This study, entitled Evaluation of the Use of Web-Based Library Management System at I Gusti Bagus Sugriwa State Hindu University, Denpasar, aims to evaluate the effectiveness, challenges, and strategic significance of WLMS implementation using the Technology Acceptance Model (TAM) and the Technology–Organization–Environment (TOE) framework. A sequential explanatory design was applied, combining quantitative surveys and qualitative interviews with 39 respondents consisting of librarians, lecturers, and students. Quantitative findings indicate very good results in perceived usefulness (PU), attitude toward use (ATU), behavioral intention (BI), and actual use (AU), while perceived ease of use (PEOU) scored in the “good” category. TOE-based analysis reveals supporting factors such as infrastructure availability, leadership commitment, and system accessibility, while inhibiting factors include unstable internet, data security concerns, insufficient training, and low digital literacy among users. These results demonstrate that although WLMS significantly improves accessibility, efficiency, and digital literacy, its sustainability requires continuous system optimization, technical support, and human resource development. This research concludes that WLMS adoption at I Gusti Bagus Sugriwa State Hindu University is largely successful but still needs strategic interventions to ensure long-term effectiveness and integration with the university’s mission of advancing Hindu knowledge and culture in the digital era
Design and Implementation of a File Encryption System Based on Object-Oriented Programming Using C++
Get PDFData security has become a critical aspect in the digital era, especially in protecting confidential files from unauthorized access. This study aims to design and implement a file encryption system based on Object-Oriented Programming (OOP) principles using the C++ programming language. The system is developed by applying the core concepts of OOP, namely encapsulation, inheritance, and polymorphism, to create a modular and reusable software architecture. The encryption process is implemented using symmetric key algorithms, allowing users to encrypt and decrypt text-based files securely. The program design follows a layered class structure, where each class handles specific functionalities such as key generation, file reading and writing, and cryptographic operations. The implementation is conducted in a console-based environment with a focus on simplicity, efficiency, and maintainability of the source code. The evaluation results show that the OOP-based approach enhances the flexibility of system modification and reduces code redundancy compared to traditional procedural programming. Furthermore, performance testing demonstrates that the encryption and decryption processes can be executed efficiently without significant latency for small to medium-sized files. This research concludes that applying OOP concepts in C++ provides a structured and scalable framework for developing secure and maintainable encryption systems, which can be further enhanced for larger-scale or GUI-based applications in the future