Publication Search

Tangerang City has the most applications in Indonesia, with 222 applications. All of these applications are supported by more than 100 servers located in the data center of the Tangerang City Communication and Information Agency. The large number of servers and applications that are managed brings up new problems in the midst of increasing complex cyber threats, especially in government data centers. One of them is how to monitor and respond quickly when there is an attack on the existing system. The implementation of a cyber security system based on Wazuh, Shuffle, and YARA is able to monitor threats in realtime and automate responses against attacks. Wazuh acts as a log-based monitoring and detection platform and behavior analysis, Shuffle is used to automate incident response through integrated workflow, and YARA is applied for signature-based malware identification. The PPDIOO (Prepare, Plan, Design, Implement, Operate, Optimize) method used in this research is used as a framework in designing and evaluating the system. From the research conducted, it is expected that Wazuh successfully monitors anomalies that occur on the server which will then be forwarded to Shuffle to automate the next steps to be taken. YARA integrated with Wazuh also successfully detects and quarantines malicious files that enter the server automatically based on the available signature list.

The rapid development of smartphone technology today creates challenges for consumers and manufacturers in determining an objective price range based on highly varied technical specifications. This study aims to implement the Random Forest algorithm in classifying smartphone price ranges into four main categories, namely low, mid-range, high, and flagship. The research method was carried out systematically through the stages of loading a dataset of 2,000 entries, exploratory data analysis (EDA) to ensure data integrity, and model training with a training and testing data split of 80:20. The results showed that the Random Forest model achieved a significant overall accuracy rate of 89%. Based on feature importance analysis, it was found that RAM capacity was the most dominant determining factor, contributing 47% to prediction accuracy, followed by battery power and screen resolution as supporting features. These findings have strategic implications for manufacturers to prioritize memory capacity upgrades in determining product pricing in the market, as well as providing guidance for consumers in assessing the fairness of a device's price based on its technical capabilities.

Per capita expenditure is an important indicator of household welfare because it reflects the economic capacity and consumption patterns of the community, as explained in Engel's Law. In regions with diverse geographical characteristics such as Papua Province, spatial analysis is needed to understand the variations in expenditure between districts/cities and the differences between urban and rural areas. This study aims to analyze the spatial distribution of per capita expenditure percentages for food and non-food items in nine districts/cities in Papua Province during the 2022–2024 period. The research data was sourced from the National Socioeconomic Survey (Susenas). The methods used included quantile-based choropleth mapping using QGIS, attribute data merging through table joins, and Pearson's correlation test to evaluate the consistency of spending patterns between years. The analysis results show that food and non-food spending patterns were relatively stable during the observation period with high correlation values (r = 0,85–0,93), although spatial variations between regions were still apparent. Mamberamo Raya Regency consistently had the highest proportion of food spending (>68%), while Jayapura City showed the lowest proportion. These findings indicate spatial disparities related to urbanization levels and economic access. Spatial visualization proved effective in revealing regional disparity patterns that were not fully apparent through conventional statistical tables and has the potential to support the formulation of more evidence-based regional development policies.

Flower disease detection is a common challenge in modern agriculture. Various factors, such as changes in leaf color, shape, petal structure, and environmental conditions, make it difficult to achieve high accuracy with conventional models. Transfer learning is an effective solution to improve model performance in image detection, especially when available data is limited. This study used several pre-trained models, namely VGG16, ResNet50, and EfficientNet-B0, to detect three types of flower diseases: black spot on roses, white powdery mildew, and leaf rust. The process included data processing, increasing the data volume, model training, and result verification. The results showed that the EfficientNet-B0 model provided the highest accuracy of 97.2%, significantly better than the CNN model created from scratch with an accuracy of 85.1%. This study proves that the transfer learning method is very effective in improving the accuracy of flower disease detection. These results confirm that transfer learning is effective for detecting plant diseases with higher accuracy, especially when the dataset is limited.

Flooding is one of the most recurrent natural disasters in Central Java Province, particularly during the rainy season. Diverse geographical characteristics, high rainfall intensity, and rapid urban development contribute to the region’s high vulnerability to flood hazards. According to the Central Java Statistics Agency, a total of 414 flood events and 407,784 affected victims were recorded between 2019 and 2021. This study aims to develop a web-based Geographic Information System (GIS) capable of mapping the spatial distribution and impact levels of floods across Central Java. The methodology includes collecting flood event data from the Central Java Statistics Agency, processing spatial data such as administrative boundary shapefiles, performing attribute integration between spatial and non-spatial datasets, and creating thematic maps using QGIS. The visualization outputs were exported into an interactive web format using the qgis2web plugin and subsequently integrated into a website developed with HTML, CSS, and JavaScript. The results show that the GIS system successfully visualizes flood-prone areas, identifies regions with high flood intensity, and enables users to explore detailed information through interactive digital maps. Additional website features—such as historical flood data, statistical summaries, and descriptive impact indicators—enhance the system's usefulness for disaster analysis. This study demonstrates the crucial role of GIS in supporting disaster mitigation, spatial planning, and policy evaluation related to flood management. Future research is recommended to incorporate more recent datasets and additional non-spatial variables such as rainfall intensity and floodwater depth to improve the system’s analytical accuracy and comprehensiveness.

The instability of the climate is becoming increasingly prominent across Southeast Asia, creating uncertainty in agricultural systems that are highly dependent on seasonal weather patterns. Indonesia, where rice remains the primary staple food, is particularly vulnerable to the effects of rising temperatures and rainfall deficits. This study applies the Seasonal Autoregressive Integrated Moving Average (SARIMA) model to predict rice production while incorporating indicators of extreme climate anomalies. Using publicly available datasets, including FAOSTAT production statistics, NOAA rainfall and temperature anomalies, and climate indices from the World Bank, this model was developed following the Box-Jenkins procedure. Among the configurations tested, the SARIMA model (1,1,1)(0,1,1)₁₂ showed the strongest performance, reflected in a MAPE of 4.62% and low RMSE values. The model indicates that significant El Niño events can reduce annual rice production by 3–7%, while wetter La Niña conditions may support production recovery. These findings highlight the importance of integrating climate-sensitive data into agricultural forecasting. The model presented here could support early warning systems, adaptive farming strategies, and long-term food security planning in Indonesia.

Communication is a fundamental human need for conveying information and ideas. However, individuals who are deaf and mute face difficulties in communicating with the broader community that does not understand sign language. This study aims to design and implement a real-time static sign language translator into speech using five flex sensors, an MPU6050 sensor, a Raspberry Pi Pico, an ADS1115 ADC module, and a DFPlayer Mini module as the audio output medium. Testing results show that the device successfully recognizes finger movements and hand orientation. The system is capable of playing audio output corresponding to recognized gestures, with the shortest latency recorded at 1.1 seconds and the longest at 2.8 seconds, achieving a detection accuracy rate of 75% based on 60 tests across 12 sign words. This device supports the translation of 12 simple sign words. The implementation demonstrates potential as an assistive communication tool, although further development is needed to improve accuracy, expand vocabulary, and conduct trials directly with deaf or mute users.

Tuberculosis (TB) is one of the infectious diseases that remains a public health problem in Indonesia, including in West Java Province, which has a large population and high mobility. This condition has the potential to increase the risk of transmission and cause variations in the distribution of cases between districts and cities. This study aims to map the distribution of TB cases across all districts and cities in West Java Province from 2022 to 2024 using a spatial analysis approach. This analysis was conducted to describe the geographical distribution of cases, identify patterns of spread, and determine areas with relatively high or low case rates. TB data was obtained from routine recording and reporting by health facilities in West Java, then integrated with population and administrative boundary data. The results of the analysis provide information on case distribution patterns between regions and trends in case changes from year to year. The findings of this study are expected to serve as a basis for local governments in formulating more targeted TB prevention and control strategies, through a focus on interventions in areas with a high case burden, as well as optimizing sustainable public health programs in West Java Province.

The use of information technology through website development has become a crucial need for MSMEs, including traditional food businesses like Siomay Batagor Bogarasa. This study aims to develop a business profile website using the Odoo-based ERP website module to strengthen brand identity and facilitate information delivery to customers. The development process follows the Software Development Life Cycle (SDLC) stages, including planning, requirement analysis, design, implementation, and testing. The developed website consists of several pages such as Home, About Us, Menu, and Contact Us, presenting information about the business history, product catalog and pricing, sales locations, and contact channels. Testing results indicate that most features function well, with responsive display, smooth navigation, active contact form, and effective social media integration. One issue was found in the Google Maps display, which has not yet accurately shown the business location. Overall, the website provides practical benefits in terms of promotion, communication, and enhancing the professionalism of the business.

The attendance recording system at PLN ULP Tanjungbalai still relies on manual, paper-based methods, resulting in delays in data recap, reduced efficiency, and a high potential for recording errors. This condition affects the accuracy of employee attendance information, which is essential for administrative activities and managerial decision-making. Based on these issues, this practical work aims to design and develop a web-based e-attendance application as a solution to enhance efficiency, processing speed, and the accuracy of attendance recapitulation. The system was developed using PHP as the programming language and MySQL as the database management system, following several stages including requirement analysis, system design using UML, and implementation of a web-based user interface. The application provides essential features such as user login, daily attendance recording, employee data management, attendance notes (permission, sickness, etc.), and automatic attendance report generation. The system is designed for two types of users—Admin and Employees—each with specific access rights. The implementation results indicate that the e-attendance application significantly improves the efficiency of attendance administration at PLN ULP Tanjungbalai. Data collection and recapitulation become faster, more structured, and less prone to errors, while also enabling administrators to monitor employee attendance in real time. Therefore, this web-based e-attendance application serves as an effective solution to support operational activities and enhance the quality of employee attendance management.

Brake discs are one of the motorcycle spare parts that function to connect and reduce speed when the motorbike is moving, although very simple, this spare part is part of the braking system and has a very important role, without braking it will endanger the driver and other road users. To determine the hardness of the motorcycle brake disc on two original brake discs and variation brake discs on the Yamaha Vision sport motorbike by conducting Rockwell hardness tests on the original and variation brake discs with 10 test variable points. The results obtained for the average value at the level of hardness on the original disc sample are 55.1 HRC and the average value on the variation disc sample is 44.6 HRC, and the conversion hardness value from the Rockwell test. For Brinell testing and tensile testing from the results of Rockwell hardness testing, the average Brinell and tensile test values ​​were converted from the test site with the average Brinell conversion results for the original disc being 575 HBS, while the average variation disc value was 417.1 HBS, the conversion tensile test value with the results on the original disc was 1626 N/mm² and the variation disc was 1427 N/mm² with 10 variable point Rockwell testing and the conversion value of the Brinell hardness test and the original disc tensile test value was still greater than the variation disc tensile test value.

While TLS 1.3 is the latest standard, TLS 1.2 remains widely implemented in many cloud infrastructures. The selection of cipher suites in TLS 1.2, particularly between AES-128-GCM and AES-256-GCM, presents a trade-off between cryptographic strength and system performance. This research aims to analyze the performance comparison of these two algorithms on an Nginx server to determine the optimal configuration for cloud storage services. The study uses a quantitative experimental method by benchmarking two scenarios: (A) Strict (AES-256-GCM), and (B) Balanced (AES-128-GCM). Performance metrics measured include Requests Per Second (RPS), Latency, and Throughput. The results show that handshake performance (RPS and Latency) is nearly identical across all scenarios. However, in large file transfer tests, the AES-128-GCM algorithm (Scenario B) achieved a throughput of 32.4 MB/s, which is 12.5% faster than AES-256-GCM (28.8 MB/s). This study concludes that AES-128-GCM provides the best balance of security and efficiency for data-intensive environments.

The Information System registration module for the Regional Taxpayer Identification Number (NPWPD) was developed using the Laravel framework and implemented by the Taxpayer Identification Agency (BPPKAD) at Kediri City. The system was designed to digitize the NPWPD registration process, which was previously done manually. This traditional approach often led to long queues, extended processing times, and, at times, errors in data entry. The new system offers several key advantages, including an online registration form that allows taxpayers to upload required documents such as photos of ID cards, business locations, and other necessary paperwork. Data validation is performed by officers to ensure accuracy, and automatic notifications are sent to taxpayers, informing them of the status of their applications. The implementation of this system has had several positive impacts, such as significantly improving the efficiency of administrative processes, reducing the manual workload for officers, and increasing transparency and accountability in public services. Moreover, it has improved customer satisfaction by providing faster, more accurate, and more responsive services. This system supports the creation of a streamlined, user-friendly, and effective method for taxpayers to register for NPWPD online, enhancing the overall quality of public sector service delivery.

Queueing systems play an important role in evaluating service performance, especially in small-scale businesses such as barbershops, where fluctuating customer arrival patterns and limited service capacity often lead to long waiting times. This study aims to analyze the performance of barbershop services using the M/M/1 queueing model and an analytical approach based on experimentally tested arrival (λ) and service (μ) rates. The model was selected because it represents a single-server system with Poisson arrivals and exponentially distributed service times, closely matching real barbershop operational characteristics. Using assumed realistic parameters, the analysis shows that when λ = 12 customers per hour and μ = 6 customers per hour, the system becomes unstable with a utilization rate (ρ) exceeding 1, indicating continuous queue growth. Further simulations with increased service rates demonstrate significant improvements: at μ = 15, the system achieves ρ = 0.8 with an average waiting time of 16 minutes, while at μ = 13, the system remains stable but experiences a long waiting time of approximately 55 minutes. These findings emphasize that barbershop performance is highly sensitive to service speed and that even small increases in μ can produce substantial improvements in queue stability and customer waiting times. The study concludes that barbershops must ensure adequate service capacity—either through optimizing service duration, improving worker efficiency, or adding servers—to maintain service quality and enhance customer satisfaction.

This study presents a high-accuracy real-time soft failure detection framework for large-scale fiber-to-the-home(FTTH) optical access network using a hybrid ensemble of Isolation Forest and One-Class Support Vector Machine (OCVSM). The proposed model was trainde and validated on a real-word multivariate performance dataset comprising more than 1.8 million samples collected at 5-minute intervals from 50 Optical Line Terminal (OLTs) and over 3,000 Optical Network Terminals (ONTs) across a five-month periode(June-October 2025). Ground-truth validation was performed using 111 confirmed network incidents in October 2025 affecting 12,990 customer. The hybrid ensemble achieved Precision 0.940, Recall 0.982, with an average detection delay of only 7.8 minutes-representing an 87.7% reduction compared to conventional manual response (63.5 minutes). The framework significantly outperforms traditional threesholding and recent ML-based methods while demonstrating practical deployability in live operational enviroments.

The rapid development of digital streaming platforms such as Netflix has generated a large volume of content data with diverse characteristics, thereby requiring effective analytical methods to understand emerging patterns and trends. This study aims to classify Netflix content into two main categories, namely movies and television shows, and to analyze genre trends and content characteristics using a data mining approach with the Naive Bayes algorithm. The dataset used in this study is the Netflix Shows dataset, consisting of 8,809 content entries, with the primary features analyzed including genre, rating, and country of production. The research process begins with data exploration and preprocessing stages, including data cleaning, handling missing values, and transforming categorical features to enable effective model construction. Subsequently, the dataset is divided into training and testing sets to objectively and systematically build and evaluate the Naive Bayes classification model. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics to assess the model’s ability to accurately distinguish between Netflix content types. The experimental results demonstrate that the Naive Bayes algorithm is able to classify Netflix content into Movie and TV Show categories with accuracy, precision, recall, and F1-score values of 100%, respectively. The confusion matrix indicates that no misclassification occurred, suggesting that genre, rating, and country of production features provide a very clear separation between content classes. These findings indicate that the Naive Bayes algorithm can achieve exceptionally high classification performance with optimal evaluation results. The results further reveal distinct differences in characteristics between movies and television shows based on genre and production attributes. Therefore, this study is expected to contribute to the development of content recommendation systems and strategic content management within the streaming industry.

Automatic motorcycle spare part sales have increased along with the high use of automatic two-wheeled vehicles in the community. To support optimal sales strategies and stock management, customer purchasing pattern analysis is required. This study uses the FP-Growth algorithm to identify association patterns between automatic motorcycle spare part products that are frequently purchased together. FP-Growth was chosen because of its ability to efficiently find frequent itemsets without the need to generate candidate itemsets as in the Apriori algorithm. Transaction data is processed to form an FP-Tree which is then extracted to find relationships between items. The analysis results show combinations of products that frequently appear together, such as brake pads and engine oil, which can be used as a basis for compiling sales packages, product placement, and product recommendations. By implementing the FP-Growth algorithm, spare part stores or workshops can improve service and efficiency in sales management.

Diabetes Mellitus is a chronic disease that requires early detection to prevent serious complications. This study aims to implement the Artificial Neural Network (ANN) algorithm with the Backpropagation method to predict the risk of diabetes. The dataset used is the Pima Indians Diabetes Dataset, consisting of 768 medical records with 8 feature attributes. This study employs the Multi-Layer Perceptron method with an architecture of 8 input neurons, two hidden layers, and 1 output neuron. Model evaluation is conducted using a Confusion Matrix to measure accuracy levels. The test results show that the model is capable of predicting diabetes diagnosis with an accuracy rate of 76.62%. Based on these results, it can be concluded that the Backpropagation algorithm is effective as an alternative method for early detection of diabetes, although further development is needed to improve the model's sensitivity to positive cases.  

Traffic congestion is a persistent challenge in urban areas in Indonesia, where increasing vehicle density creates the need for intelligent traffic monitoring systems. This study aims to develop a real-time vehicle parking system using the YOLOv8 object detection model to provide efficient traffic analysis from live CCTV broadcasts and recorded videos. This study uses a quantitative experimental approach with the implementation of the YOLOv8m model using the Ultralytics library in Python, tested on data collected from CCTV cameras A TCS Dishub Medan and additional footage from mobile devices. Vehicles are detected and counted in two directions up (Up) and down (Down) using virtual detection lines on the video frame. The system performance is evaluated by automatic detection counting with manually recorded ground truth data. The results show that on live CCTV broadcasts, the YOLOv8m model achieves an average precision of 98.96%, a recall of 96.59%, and an F1 score of 97.74% for upstream traffic, while for downstream traffic it achieves 100% precision, 95.64% recall, and an F1 score of 97.730/0. On the other hand, on high-quality recorded videos, all performance metrics achieve 100%, indicating perfect detection accuracy. These findings confirm the effectiveness of YOLOv8 in real-time traffic monitoring, but also indicate that video quality and stream stability affect detection performance. In conclusion, the developed system shows strong potential to support smart city traffic management solutions. Future research should focus on performance optimization under low-resolution live streaming conditions to improve accuracy in practical applications.

The scarcity of real-world data in Air-Conditioning (AC) fault diagnosis necessitates the use of synthetic data; however, rule-based synthetic datasets often suffer from a significant sim-to-real domain gap. To address this, we propose a Model-Data Coevolution (MDC) framework that employs a Simulated Annealing (SA) controller to optimize augmentation parameters. We introduce a novel technique, Stochastic Feature Decoupling (SFD), which applies independent noise to raw and derived features, contrasting it with traditional Logically-Consistent Augmentation (LCA). Empirical results show that SFD significantly outperforms LCA, achieving a weighted F1-score of 0.93 and increasing NORMAL class recall to 82%. We demonstrate that by breaking deterministic feature links, SFD acts as a robust regularizer, utilizing "physically impossible" data to enhance generalization in complex real-world environments.