Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Design and Implementation of an IoT-Based Smart Drip Irrigation System Using Takagi-Sugeno Fuzzy Logic for Melon Cultivation
No full textThe melon plant (Cucumis melo L.), a species of the Cucurbitaceae family, requires precise water management to support optimal growth. This study developed an Internet of Things (IoT)-based innovative irrigation system employing Takagi-Sugeno fuzzy logic to regulate water supply for melon cultivation in a greenhouse. The system integrates a capacitive soil moisture sensor and a DS18B20 temperature sensor, both connected to an ESP8266 microcontroller, which controls a solenoid valve used in the drip irrigation method. Sensor data are transmitted in real-time to Firebase Realtime Database (cloud platform) for monitoring through a web-based interface. The solenoid valve opening duration ranges from 0 to 720 seconds per irrigation session, dynamically adjusted according to soil moisture and temperature inputs. Experimental results demonstrate that the proposed system effectively maintains soil moisture within the optimal range of 60%–80%. However, plant growth evaluation indicates that the system has not fully promoted healthy development, particularly in plant height and leaf width, likely due to additional factors such as soil conditions, humidity, and nutrient availability. Despite these limitations, the proposed smart irrigation system shows strong potential for further refinement to enhance water efficiency and support sustainable melon cultivation.The melon plant (Cucumis melo L.), a species of the Cucurbitaceae family, requires precise water management to support optimal growth. This study developed an Internet of Things (IoT)-based innovative irrigation system employing Takagi-Sugeno fuzzy logic to regulate water supply for melon cultivation in a greenhouse. The system integrates a capacitive soil moisture sensor and a DS18B20 temperature sensor, both connected to an ESP8266 microcontroller, which controls a solenoid valve used in the drip irrigation method. Sensor data are transmitted in real-time to Firebase Realtime Database (cloud platform) for monitoring through a web-based interface. The solenoid valve opening duration ranges from 0 to 720 seconds per irrigation session, dynamically adjusted according to soil moisture and temperature inputs. Experimental results demonstrate that the proposed system effectively maintains soil moisture within the optimal range of 60%–80%. However, plant growth evaluation indicates that the system has not fully promoted healthy development, particularly in plant height and leaf width, likely due to additional factors such as soil conditions, humidity, and nutrient availability. Despite these limitations, the proposed smart irrigation system shows strong potential for further refinement to enhance water efficiency and support sustainable melon cultivation
An Ensemble Learning Approach for Sentiment Analysis of Maxim Application Reviews Using SVM, KNN, and Random Forest
No full textThe development of online transportation applications such as Maxim has increased the need for sentiment analysis to understand user opinions from reviews on the Google Play Store. The main challenges in this analysis are language diversity, variations in writing style, and data imbalance, which affect model accuracy. This study aims to evaluate the performance of the Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest (RF) algorithms, as well as ensemble approaches through the Voting Classifier and Combined Classifier, in sentiment analysis of Maxim app reviews. The dataset consists of 2,851 Indonesian-language reviews collected through web scraping from the Google Play Store in 2025. Sentiment labels were automatically determined based on user ratings, where ratings of 4–5 were categorized as positive and ratings below 4 as negative, with an initial distribution of 2,295 positive and 556 negative reviews before balancing using SMOTE–Tomek Links. Preprocessing steps included case folding, tokenization, stopword removal, and stemming using Sastrawi, while feature weighting was performed with unigram TF-IDF. The Combined Classifier merged the probability scores from the SVM, KNN, and RF models to produce the final prediction. Evaluation was conducted using 5-Fold Cross Validation with accuracy, precision, recall, F1-score, and ROC-AUC as evaluation metrics. The results show that RF and the Combined Classifier achieved the best performance with 85% accuracy, 87% precision, 85% recall, 86% F1-score, and 0.91 ROC-AUC, while SVM and the Voting Classifier ranked in the middle and KNN ranked the lowest. These findings confirm that ensemble learning, particularly the Combined Classifier, effectively improves the accuracy and stability of review classification compared to individual methods
Enhancing the Predictive Accuracy of Corrosion Inhibition Efficiency Using Gradient Boosting with Feature Engineering and Gaussian Mixture Model
No full textPrediction The development of Quantitative structure property relationship (QSPR) models for predicting corrosion inhibition efficiency (IE) often faces challenges due to small datasets, which heightens the risk of overfitting and results in less reliable performance assessments. This research creates an entirely leakage-free modeling framework by combining per-fold preprocessing, augmentation of training-only data, and rigorous Leave-One-Out Cross-Validation (LOOCV). A set of 20 pyridazine derivatives was evaluated using 12 quantum-chemical descriptors, including HOMO, LUMO, ΔE, dipole moment, electronegativity, hardness, softness, and the electron-transfer fraction. An initial assessment showed that all baseline models lacking augmentation Gradient Boosting, Random Forest, SVR, and XGBoost demonstrated limited predictive power (R² < 0.20), revealing the dataset\u27s inherently low information complexity.To enhance representation in the feature space, a multi-scale Gaussian Mixture Model (GMM) was used to generate chemically valid synthetic samples, with all components trained solely on the training subset from each LOOCV fold. This strategy consistently improved model performance. The two most successful configurations, XGBoost + GMM v2 and Random Forest + GMM v3, reached R² values of 0.4457 and 0.4108, respectively, along with significant decreases in RMSE, MAE, and MAPE. These findings illustrate that GMM-based generative augmentation effectively captures multicluster structures within the descriptor space while expanding the chemical variability domain in a controlled way.While the resulting R² values remain inadequate for high-precision quantitative predictions, the proposed methodology provides a solid basis for early-stage evaluation of corrosion inhibitors in situations with limited data. Future research will aim to integrate advanced DFT-derived descriptors, molecular graph representations, and tests against larger external datasets to enhance model generalizability
Implementation and Comparative Analysis of CNN and Transfer Learning Models (EfficientNetB0, MobileNetV2, and ResNet50) for Rice Leaf Disease Detection Based on Digital Images
No full textRice leaf diseases significantly reduce agricultural productivity, making early and accurate detection essential, particularly in rice-producing regions such as Indonesia. This study proposes an automated rice leaf disease detection system based on Convolutional Neural Networks (CNN) and transfer learning. The dataset, obtained from the Mendeley Data Repository, consists of 6,889 images classified into eight categories: Bacterial Leaf Blight, Brown Spot, Healthy Rice Leaf, Leaf Blast, Leaf Scald, Narrow Brown Leaf Spot, Rice Hispa, and Sheath Blight. The dataset was divided into 70% training, 15% validation, and 15% testing. A baseline CNN model and three pre-trained models—EfficientNetB0, MobileNetV2, and ResNet50—were evaluated using accuracy, precision, recall, F1-score, and confusion matrix analysis. The baseline CNN achieved a test accuracy of 48.26%, while EfficientNetB0 achieved 58.41%. In contrast, MobileNetV2 and ResNet50 demonstrated significantly better performance, with test accuracies of 79.98% and 76.60%, respectively. MobileNetV2 exhibited the most balanced performance across all classes, showing superior generalization capability and computational efficiency. The best-performing model was integrated into a Streamlit-based application, enabling real-time rice leaf disease detection through image upload. The results confirm that transfer learning substantially improves classification accuracy and robustness compared to conventional CNNs. This study highlights the potential of lightweight deep learning models for practical implementation in smart agriculture systems and provides a reliable solution for automated rice disease detection in real-world conditions
Optimization of IndoBERT for Sentiment Analysis of FOMO on Social Media Through Fine-Tuning and Hybrid Labeling
No full textThe rapid growth of social media in Indonesia has given rise to social phenomena such as Fear of Missing Out (FOMO). Expressions of FOMO on platforms like X (previously Twitter) often written informally, filled with abbreviations, slang, and emotional nuances, posing challenges for traditional Natural Language Processing (NLP) methods. This research aims to develop an optimized sentiment classification model for FOMO-related posts by fine-tuning the IndoBERT architecture and applying comprehensive data enhancement strategies. The study introduces three key innovations: (1) systematic text normalization to handle informal expressions, (2) a hybrid labeling framework combining automated model prediction, lexicon-based validation, and manual annotation to construct high-quality ground-truth data, and (3) hyperparameter tuning using both GridSearchCV for traditional machine learning models and Bayesian Optimization (Optuna) for deep learning models to maximize performance. The experimental results demonstrate that the optimized IndoBERT achieved superior performance with an Accuracy of 94.50%, F1-Score of 94.52%, and Macro AUC of 0.987. These results significantly surpass comparative models, including BiLSTM (Accuracy 86.60%), Support Vector Machine (88.06%), and Naive Bayes (80.73%). These results confirm that integrating hybrid labeling and fine-tuned IndoBERT significantly enhances sentiment classification performance. The findings contribute to developing reliable sentiment analysis systems for detecting social anxiety dynamics and computational social science research in Indonesian contexts.The rapid growth of social media in Indonesia has given rise to social phenomena such as Fear of Missing Out (FOMO). Expressions of FOMO on platforms like X (previously Twitter) often written informally, filled with abbreviations, slang, and emotional nuances, posing challenges for traditional Natural Language Processing (NLP) methods. This research aims to develop an optimized sentiment classification model for FOMO-related posts by fine-tuning the IndoBERT architecture and applying comprehensive data enhancement strategies. The study introduces three key innovations: (1) systematic text normalization to handle informal expressions, (2) a hybrid labeling framework combining automated model prediction, lexicon-based validation, and manual annotation to construct high-quality ground-truth data, and (3) hyperparameter tuning using both GridSearchCV for traditional machine learning models and Bayesian Optimization (Optuna) for deep learning models to maximize performance. The experimental results demonstrate that the optimized IndoBERT achieved superior performance with an Accuracy of 94.50%, F1-Score of 94.52%, and Macro AUC of 0.987. These results significantly surpass comparative models, including BiLSTM (Accuracy 86.60%), Support Vector Machine (88.06%), and Naive Bayes (80.73%). These results confirm that integrating hybrid labeling and fine-tuned IndoBERT significantly enhances sentiment classification performance. The findings contribute to developing reliable sentiment analysis systems for detecting social anxiety dynamics and computational social science research in Indonesian contexts
Optimizing Layout for Material Handling Cost Reduction Using Blocplan and FlexSim: A Case Study in Screen Printing Production
No full textSmall-scale screen printing businesses often face challenges related to limited production space and high material handling costs, which reduce production efficiency. This research aims to design an efficient facility layout and reduce material handling costs at Usaha Sablon XYZ by using the Systematic Layout Planning (SLP) method combined with the BLOCPLAN algorithm and FlexSim. The methodology involves direct observation, interviews, initial layout measurements, and analysis using tools such as Activity Relationship Chart (ARC), Form to Chart (FTC), and Operation Process Chart (OPC). The proposed layouts produce two alternatives, with Layout Proposal 2 showing the best efficiency, reducing total material handling cost from Rp 11.999.965 to Rp 7.734.185,65 a cost reduction of 35.55%. The results indicate that BLOCPLAN and Flexsim was effective within the context of this case study in generating layout alternatives that reduce material handling cost
PENGARUH PENERAPAN GREEN ACCOUNTING, ENVIRONMENTAL PERFORMANCE DAN LIKUIDITAS TERHADAP PROFITABILITAS PERUSAHAAN YANG TERINDEKS DI IDX80: PENGARUH PENERAPAN GREEN ACCOUNTING, ENVIRONMENTAL PERFORMANCE DAN LIKUIDITAS TERHADAP PROFITABILITAS PERUSAHAAN YANG TERINDEKS DI IDX80
No full textIncreasing environmental concerns and the urgency to achieve sustainability have prompted companies to strengthen their environmental impact mitigation and enhance transparency. However, the profitability of several LQ45 companies declined in 2023. The inconsistency of literature findings regarding the influence of Green accounting, environmental performance, and liquidity on profitability is a core issue. The objective of this research is to analyze how the implementation of green accounting, environmental performance, and liquidity affects the profitability of companies indexed in IDX80 during the 2019-2023 period. The research method used is panel data regression analysis, with data processed using Eviews 13 software. The study\u27s findings reveal that green accounting does not have a positive effect on profitability. Conversely, environmental performance and liquidity are found to have a positive effect on profitability. Theoretically, this study enriches the literature on the relationship between environmental practices, liquidity, and financial performance in the Indonesian capital market. Practically, company management can use these findings as a guide to improve environmental performance and manage liquidity for profitability and competitiveness. Investors are advised to consider environmental performance as a key determinant in investment decisions, and regulators can evaluate green accounting policies to encourage more tangible financial impacts.Isu-isu lingkungan yang meningkat dan urgensi untuk mencapai keberlanjutan telah mendorong perusahaan untuk memperkuat mitigasi dampak lingkungan dan meningkatkan transparansi, namun profitabilitas sejumlah perusahaan LQ45 menurun pada tahun 2023. Inkonsistensi temuan literatur mengenai pengaruh green accounting, environmental performance, dan likuiditas terhadap profitabilitas menjadi isu pokok. Tujuan penelitian ini adalah untuk menganalisis bagaimana penerapan green accounting, environmental performance, dan likuiditas memengaruhi profitabilitas perusahaan yang terindeks IDX80 periode 2019-2023. Metode penelitian yang digunakan adalah analisis regresi data panel, pengolahan data menggunakan software Eviews 13. Hasil penelitian ditemukan bahwa green accounting tidak berpengaruh positif terhadap profitabilitas. Sebaliknya, environmental performance dan likuiditas ditemukan berpengaruh positif terhadap profitabilitas. Secara teoretis, studi ini memperkaya literatur mengenai hubungan antara praktik lingkungan, likuiditas, dan kinerja keuangan di pasar modal Indonesia. Secara praktis, manajemen perusahaan dapat menggunakan temuan ini sebagai panduan untuk meningkatkan environmental performance dan mengelola likuiditas demi profitabilitas dan daya saing. Investor disarankan mempertimbangkan environmental performance sebagai penentu investasi, dan regulator dapat mengevaluasi kebijakan green accounting untuk mendorong dampak finansial yang lebih nyata
The Role of Trust in the Intention to Continue Using Mobile Payment Applications in Indonesia: A Case Study of Gen Z in Batam City
No full textThis study aims to evaluate how trust influences the intention of Generation Z in Batam City to continue using mobile payment applications. Using a quantitative approach, data were collected from 348 respondents through a survey and analyzed using the SEM-PLS method. The analysis included tests for validity (AVE, loadings), reliability (Cronbach’s Alpha, Composite Reliability), model fit (SRMR), path analysis using bootstrapping, and moderation testing using MICOM and Multi-Group Analysis (MGA). The findings indicate that security and application reputation have a positive and significant effect on building user trust. In contrast, mobility and customization do not show a significant impact. Trust is proven to significantly reduce risk perception and enhance the intention to continue using the application. On the other hand, risk perception negatively affects the sustainability of application usage. The moderation analysis using MICOM and Multi-Group Analysis reveals that gender does not significantly moderate the relationship between antecedents and trust. These findings underscore the crucial role of security and reputation in cultivating user loyalty among Gen Z mobile payment users
Feasibility Analysis of the Development Plan of Air Putih-Selat Baru Road from Traffic Perspectives
No full textAlong with the development of the population and according to the spatial layout of the Bengkalis Regency, a road plan is needed that can facilitate movement from Bengkalis District to Bantan District. Based on this condition, the Public Works and Spatial Planning Office of Bengkalis Regency plans to build the Air Putih - Selat Baru Poros Road. However, before carrying out the road construction work, it is necessary to assess the feasibility of the planned road construction using one of the study stages, namely a feasibility study. The method used in this study is the with and without method in terms of traffic. In the feasibility analysis from a Traffic point of view, the Degree of Saturation (DJ) indicator is used. After conducting a feasibility analysis, the results show that the degree of saturation of the existing road is still below the standard value (0.85), namely 0.15, so it can be said that it is not feasible in terms of traffic
Land Cover Modeling in 2034 in Waiheru Watershed, Ambon City, Indonesia Using CA-Markov
No full textThe increasing demand for land due to population growth and community activities in Ambon City certainly has an impact on land cover changes in the Waiheru Watershed. Therefore, it is important to model future land cover as a sustainable environmental planning material. This study aims to analyze land cover changes in 2014, 2019, 2024 and model land cover in 2034 with the Cellular Automata Markov Chain (CA-Markov) approach. This method integrates historical land cover data and factors driving land use change, elevation, slope, distance from road, distance from river, population, distance from point of interst. The results show a trend of conversion of mixed agricultural land into residential areas, which certainly has the potential to exacerbate ecosystem vulnerability in the Waiheru watershed. Settlement land cover continues to increase in area, namely 68.78 ha in 2014, 77.73 ha in 2019, 96.72 ha in 2024 and the modeling results in 2034 show that settlement land has an area of 138.65 ha, this is in contrast to the forest area which has decreased due to the expansion of population settlements.These findings are expected to provide insights for policy makers and urban planners in formulating sustainable land management strategies, as well as maintaining a balance between development needs and environmental conservation in the Waiheru watershed