Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Eye Disease Classification Using EfficientNet-B0 Based on Transfer Learning
Get PDFThis study focuses on developing and evaluating a deep learning approach employing EfficientNet-B0 based on transfer learning to classify retinal fundus images into four categories: Cataract, Diabetic Retinopathy, Glaucoma, and Normal. The model was trained using a retinal image dataset and demonstrated stable training performance, indicated by a consistent decrease in both training and validation loss without signs of overfitting. The training accuracy reached 92%, while the validation accuracy ranged between 94–95%. Model performance evaluation using a confusion matrix and classification report showed excellent classification results, particularly for the Diabetic Retinopathy class, with an F1-Score of 0.98. The Cataract and Normal classes also achieved high performance, with F1-Scores of 0.94 and 0.92, respectively. However, classification accuracy slightly declined for the Glaucoma class, which experienced some misclassification with the Normal class. Overall, the model achieved a classification accuracy of 94% on the test dataset, indicating good generalization capability. These findings suggest that the model holds strong potential for implementation in automated medical image-based diagnostic support systems. Nonetheless, performance improvement in classes with relatively higher misclassification rates is still required to ensure model reliability in clinical practice.This study focuses on developing and evaluating a deep learning approach employing EfficientNet-B0 based on transfer learning to classify retinal fundus images into four categories: Cataract, Diabetic Retinopathy, Glaucoma, and Normal. The model was trained using a retinal image dataset and demonstrated stable training performance, indicated by a consistent decrease in both training and validation loss without signs of overfitting. The training accuracy reached 92%, while the validation accuracy ranged between 94–95%. Model performance evaluation using a confusion matrix and classification report showed excellent classification results, particularly for the Diabetic Retinopathy class, with an F1-Score of 0.98. The Cataract and Normal classes also achieved high performance, with F1-Scores of 0.94 and 0.92, respectively. However, classification accuracy slightly declined for the Glaucoma class, which experienced some misclassification with the Normal class. Overall, the model achieved a classification accuracy of 94% on the test dataset, indicating good generalization capability. These findings suggest that the model holds strong potential for implementation in automated medical image-based diagnostic support systems. Nonetheless, performance improvement in classes with relatively higher misclassification rates is still required to ensure model reliability in clinical practice
A Banana Disease Detection Using MobileNetV2 Model Based on Adam Optimizer
Get PDFThe main objective of this study is to develop a deep learning-based disease detection system for banana plants using the MobileNetV2 architecture through a comprehensive comparison with VGG16. This study utilizes a dataset of 3,653 images categorized into 12 classes, including Aphids, Bacterial Soft Rot, Bract Mosaic Virus, Cordana, Insect Pest, Moko, Panama, Fusarium Wilt, Black Sigatoka, Yellow Sigatoka, Pestalotiopsis, and healthy specimens. The methodological framework includes architecture comparison, data balancing, preprocessing techniques, and performance evaluation. The dataset was divided with a distribution ratio of 75% for training, 15% for validation, and 10% for testing. Comparative analysis shows excellent performance of MobileNetV2 with an accuracy of 96.21% compared to 90.15% for VGG16, while maintaining a significantly smaller model size of 10.0 MB compared to 57.8 MB for VGG16. Statistical validation through the McNemar test confirms significant superiority with a p-value of 0.008. The findings of this study contribute positively to the development of agricultural technology, particularly in the development of automated systems for disease detection in banana plants
Decision Support System for Sunscreen Selection Based on Facial Skin Concerns Using the Analytic Network Process
Get PDFExposure to ultraviolet (UV) radiation is one of the primary causes of premature skin aging and various facial skin problems. However, selecting an appropriate sunscreen product remains challenging due to limited consumer knowledge and the overlapping nature of facial skin concerns. This study proposes a decision support model using the Analytic Network Process (ANP) to determine the most suitable sunscreen product based on six common skin problems: acne-prone skin, very dry skin, outdoor-induced dullness, aging, hyperpigmentation and acne scars, and general dullness. These criteria were derived from literature and validated by a certified skincare expert. Nine sunscreen alternatives from the Wardah brand—chosen due to their wide usage in the Indonesian market and varying SPF, PA levels, and formulations—were evaluated. Expert judgment was used in pairwise comparisons, with Consistency Ratio (CR) used to ensure reliability. The ANP model was developed using unweighted, weighted, and limit supermatrices. Results showed that Wardah UV Shield Aqua Fresh Sunscreen Serum SPF 50 PA++++ had the highest global priority score. A prototype web-based system was built using PHP and MySQL to deliver personalized sunscreen recommendations. The novelty of this study lies in its integration of expert dermatological insights and the use of ANP to address interrelated skin concerns, which are rarely explored in prior skincare decision support research
Early Detection of Type 2 Diabetes Using C4.5 Decision Tree Algorithm on Clinical Health Records
Get PDFType 2 Diabetes is a chronic metabolic disorder marked by elevated blood glucose levels. It is the most prevalent form of diabetes in society, commonly triggered by poor lifestyle habits and hereditary factors. If left unmanaged, the disease can lead to serious complications such as hypertension and other chronic conditions. Therefore, early detection plays a critical role in minimizing long-term impacts and promoting healthier behavioral changes. This research focuses on classifying Type 2 Diabetes using clinical data with the C4.5 Decision Tree algorithm. The dataset encompasses attributes including gender, age, height, weight, waist circumference, BMI, systolic and diastolic blood pressure, respiratory rate, and pulse rate. The model was evaluated under two scenarios: without data balancing and after applying the SMOTE technique for balancing. In the first scenario, the best performance was achieved with a training-testing split of 80:20, resulting in an F1 Score of 67.76%. However, the performance varied across different data proportions. In contrast, the second scenario showed more consistent results, with the 60:40 split yielding the highest F1 Score of 66.67%. These findings suggest that SMOTE effectively reduces bias toward the majority class and enhances sensitivity to the minority class. Therefore, data balancing is a crucial step in developing a reliable classification model for Diabetes Mellitus diagnosis
Application of Artificial Neural Network (MLP) for Multivariate Analysis of Stunting Causes in Indonesia
Get PDFStunting is a major public health challenge in Indonesia, primarily caused by prolonged malnutrition and recurrent infections during the First 1,000 Days of Life. This study utilizes the Multi-Layer Perceptron (MLP) neural network model to predict stunting, offering a new dimension in the analysis of complex data and identification of patterns influencing stunting. With its capabilities, the MLP model provides higher precision in detecting contributing factors to stunting. The evaluation results of the model show RMSE of 0.7231, MAE of 3.0313, and an R² value of 0.9463. The Food Security Index (IKP), feature X9, had the highest feature importance, followed by X5 (Lack of Clean Water) and X1 (NCPR). This study presents a novel approach to predicting stunting percentages and offers more objective insights to support evidence-based and effective health policies aimed at reducing stunting prevalence in Indonesia.Stunting is a major public health challenge in Indonesia, primarily caused by prolonged malnutrition and recurrent infections during the First 1,000 Days of Life. This study utilizes the Multi-Layer Perceptron (MLP) neural network model to predict stunting, offering a new dimension in the analysis of complex data and identification of patterns influencing stunting. With its capabilities, the MLP model provides higher precision in detecting contributing factors to stunting. The evaluation results of the model show RMSE of 0.7231, MAE of 3.0313, and an R² value of 0.9463. The Food Security Index (IKP), feature X9, had the highest feature importance, followed by X5 (Lack of Clean Water) and X1 (NCPR). This study presents a novel approach to predicting stunting percentages and offers more objective insights to support evidence-based and effective health policies aimed at reducing stunting prevalence in Indonesia
Exploration of Machine Learning Algorithms and Class Imbalance Handling on Plant Disease Detection
No full textPlant leaf diseases pose a significant threat to agricultural productivity, necessitating accurate and efficient identification systems for timely intervention. This study proposes an approach that leverages deep feature extraction using a pretrained ResNet50 model combined with traditional machine learning algorithms to recognize 38 types of plant leaf diseases. Each image was transformed into a 2048-dimensional feature vector, followed by normalization and dimensionality reduction using Principal Component Analysis (PCA). To mitigate the issue of class imbalance in the dataset, random under-sampling was applied at the feature level to ensure equal representation across all classes. Eleven machine learning models were trained and evaluated using 5-fold cross-validation, with performance assessed through accuracy, precision, recall, F1-score, and ROC AUC score. Among the evaluated models, the Support Vector Machine (SVM) achieved the highest accuracy of 99.63%, followed by Logistic Regression at 97.33%, and LightGBM at 96.25%. These models demonstrated strong generalization capabilities in multiclass settings, while simpler classifiers like AdaBoost and Decision Tree yielded lower performance. A comparative analysis of training and test accuracy further highlighted model robustness and overfitting tendencies. The findings emphasize the potential of combining pretrained convolutional neural networks for feature extraction with conventional classifiers to address complex agricultural classification tasks. Future work may explore the inclusion of healthy leaf samples, alternative CNN architectures, and deployment in real-time diagnostic tools to support precision farming and improve crop health monitoring
The Role of Centralization, Lean Operation, Risk Management Enhancement, and TB Organization Structure Implementation in Corporate Banking Operational Efficiency
Get PDFThis study investigates the synergistic effects of operational centralization, risk management enhancement, and the Transaction Banking (TB) organization model in improving banking operational efficiency through lean operation principles at the SENTINEL Bank Jakarta Branch. Root cause analyses, including 5 Whys, Fishbone, and Pareto methods, identified key inefficiencies such as decentralization, unclear roles, lack of risk awareness, and fragmented governance. A mixed-methods approach was employed, combining Structural Equation Modeling-Partial Least Squares (SEM-PLS) analysis of employee surveys (n = 200) with workload benchmarking of transactional and Full-Time Equivalent (FTE) data. The findings reveal that operational centralization significantly enhances operational efficiency (STDEV = 0.066, T = 7.056, p = 0.000) by reducing non-value-added activities, standardizing workflows, and enabling over 70% cost savings. Lean operation principles also demonstrated a significant positive impact (STDEV = 0.249, T = 3.539, p = 0.000), with centralization and TB reorganization further strengthening lean practices. However, risk management enhancements, while positive, had no significant direct effect on operational efficiency (STDEV = 0.069, T = 0.827, p = 0.408), suggesting a more supportive rather than primary role. Similarly, the TB organizational structure’s direct impact on efficiency was statistically insignificant (STDEV = 0.240, T = 1.834, p = 0.067), indicating that structural changes must be accompanied by process optimization to realize full efficiency gains. This study contributes to the discourse on corporate banking efficiency by quantifying the interdependencies among structural reforms, lean operations, and risk management practices in addressing operational inefficiencies
Generation Perceptions and Interests of Millennials in Zakat Digitally in Medan City
Get PDFThis study aims to analyze the relationship between perceptions and interests in digital zakat payments among millennials in Medan City. This is based on the rapid development of technology, finance, and zakat digitalization, which significant improvements have not yet matched in national zakat collection realization. This study employs a quantitative approach with a descriptive correlational design. The sample consists of 85 millennial respondents living in Medan City who have experience or knowledge of digital zakat platforms. Data were collected through a closed-ended questionnaire using a Likert scale and analyzed using validity, reliability, Pearson correlation, and linearity tests. The results indicate that millennials\u27 perceptions of digital zakat fall into the positive category, with an average score of 65.92. Meanwhile, their interest is also classified as high, with an average of 48.16. The correlation analysis shows a positive and significant relationship between perceptions and interests in digital zakat (r = 0.563; p < 0.001). This confirms that stronger perceptions of the convenience, security, and benefits of digital zakat led to higher interest in paying zakat through digital channels. This research makes a significant contribution to designing more inclusive, educational, and relevant digital zakat collection strategies for millennials in the digital era
OPTIMALISASI SUMBER DAYA DAN TEKNOLOGI DALAM PROGRAM PEMBERDAYAAN MASYARAKAT DESA PULAU JEMARE
Get PDFThis paper discusses in depth the implementation of the Village Community Empowerment Program (P2MD) carried out on Jemare Island. The main focus of this program is the application of appropriate technology, which includes public street lighting and a water pump system, both powered by solar panels. The main objective of this initiative is to improve the quality of lighting and the provision of clean water on the island, two important aspects that are greatly needed by the local residents. Through a series of intensive observations and discussions with the local residents, the community service team was able to identify and solve various problems faced by the residents. As a result of these efforts, four street lights have been successfully installed and operated, and the water pump system has been successfully connected to the water pump machine and water storage tank. It is hoped that this initiative can have a significant positive impact on the quality of life of the residents of Jemare Island, support their business activities and economy, and provide a solution to the uneven distribution of electricity by the state electricity company (PLN). This solution is practical because it utilizes renewable energy in the form of sunlight, which is abundantly available on Jemare Island.This paper discusses in depth the implementation of the Village Community Empowerment Program (P2MD) carried out on Jemare Island. The main focus of this program is the application of appropriate technology, which includes public street lighting and a water pump system, both powered by solar panels. The main objective of this initiative is to improve the quality of lighting and the provision of clean water on the island, two important aspects that are greatly needed by the local residents. Through a series of intensive observations and discussions with the local residents, the community service team was able to identify and solve various problems faced by the residents. As a result of these efforts, four street lights have been successfully installed and operated, and the water pump system has been successfully connected to the water pump machine and water storage tank. It is hoped that this initiative can have a significant positive impact on the quality of life of the residents of Jemare Island, support their business activities and economy, and provide a solution to the uneven distribution of electricity by the state electricity company (PLN). This solution is practical because it utilizes renewable energy in the form of sunlight, which is abundantly available on Jemare Island.
Keywords: Village Community Empowerment Program (P2MD), Appropriate technology, Solar panels, Jemare Island
MONITORING DAN KONTROL KUALITAS AIR KOLAM IKAN BERBASIS IOT
Get PDFIkan nila merupakan salah satu jenis ikan budidaya yang menjadi komoditas ekspor. Kelompok Tani ikan nila di Kecamatan Teluk Sebong Kabupaten Bintan merupakan salah satu kelompok tani yang melakukan budi daya ikan nila di Provinsi Kepulauan Riau, Namun mereka selalu mengalami kesulitan dalam menjaga kualitas air kolam budidaya secara optimal. Beberapa parameter fisik yang harus diamati untuk menggambarkan kualitas air antara lain adalah derajat keasaman (pH), kekeruhan air (total dissolved solid), suhu dan kandungan oksigen dalam air. Selama ini, pemantauan air kolam dilakukan secara manual yang sangat tergantung pada kehadiran petani, sehingga menimbulkan risiko keterlambatan penanganan. Keterlambatan penangangan perubahan kualitas air ini menyebabkan tidak optimalnya hasil panen. Untuk meningkatkan jumlah produksi ikan diperlukan pemantauan kualitas air yang real time dan akurat. Hal ini disebabkan kondisi fisik air dapat berubah dalam waktu yang relatif cepat, terutama karena adanya polutan seperti paparan limbah dan sisa makanan. Kegiatan pengabdian ini bertujuan untuk mengatasi permasalah tersebut dengan cara memfabrikasi sebuah teknologi yang mampu memonitoring dan mengontrol kualitas air berbasis Internet of Things (IoT). Sistem yang dirancang mampu memantau parameter pH, kekeruhan (TDS), oksigen terlarut (DO), suhu, dan turbidity secara real-time dan memberikan opsi kontrol otomatis berbasis aplikasi. Kelompok Tani Ikan Nila Kecamatan Teluk Sebong Kabupaten Bintan dijadikan Mitra dalam Kegiatan Pengabdian ini. Diharapkan dengan adanya sistem ini, petani ikan tidak lagi membutuhkan waktu yang lama dalam pemantauan dan persentase kematian ikan yang ada dikolam dapat diperkecil. Setelah alat ini dipasang dikolam, dan berfungsi sebagai mana mestinya, pemilik kolam merasa sangat terbantu dalam melaksanakan usahanyaIkan nila merupakan salah satu jenis ikan budidaya yang menjadi komunitas yang diekspor. Untuk menaikkan nilai eksportnya maka produksi ikan nila harus lebih ditingkatkan. Salah satu solusi agar produksi dapat ditingkatkan melalui rekayasa lingkungan. Umumnya petani ikan melakukan pemantauan kualitas air (pH, kadar oksigen terlarut dan kekeruhan air) secara manual. Pada kegiatan pengabdian ini akan dibuat sistem yang dapat memonitoring dan mengontrol kualitas air pada kolam secara otomatis berbasis IoT. Diharapkan dengan adanya sistem pemantauan secara otomatis ini, petani ikan tidak lagi membutuhkan waktu yang lama dalam pemantauan dan persentase kematian ikan yang ada dikolam dapat diperkecil. Setelah alat ini dipasang dikolam, dan berfungsi sebagai mana mestinya, pemilik kolam merasa sangat terbantu dalam melaksanakan usahanya