Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Determinants of Sustainability Report With Company Size as Moderation
Full text linkSustainability reports are important because they relate to the economic, environmental and social impacts caused by a company\u27s activities. This study aims to prove the effect of leverage, liquidity, institutional ownership, managerial ownership on sustainability reports with company size as a moderating variable and profitability as a control variable. This study used purposive sampling, resulting in a sample of 29 Consumer Goods companies listed on the IDX for the 2018-2022 period. The study found that the sustainability reports of Consumer Goods companies are significantly and negatively affected by leverage. Liquidity and managerial ownership do not affect sustainability reporting, and sustainability reporting is positively and significantly affected by institutional ownership. In addition, the relationship between sustainability reports and leverage is supported by the size of the company. However, in consumer goods companies, the relationship between institutional ownership, managerial ownership, and liquidity in sustainability reports is not reinforced by the size of the company
Optimasi Level of Detail dan Occlusion Culling dalam Game Aaron Lost in the Jungle
Full text linkThis research analyzes the effectiveness of Culling Occlusion optimization techniques andLevel of Detail (LOD) optimization techniques in improving the performance of non-optimized 3D games.optimized. Using quantitative research methods with an experimental approach.experimental approach. Culling Occlusion is a technique that hides objects that are not visible to the player.that are not visible to the player, while LOD is a technique thatLOD is a technique that displays a version of the object with reduced detail based on the distanceThe results show that by using the LOD technique managed to increase the FPS by 107%, while using theOcclusion Culling technique only marginally improved by 19%. This finding indicates that while both techniques are effective in optimizing performance, the LOD technique provides aboth techniques are effective in optimizing performance, the LOD technique provides a more significant improvementmore significant.Penelitian ini menganalisis efektivitas teknik optimasi Culling Occlusion danLevel of Detail (LOD) dalam meningkatkan performa game 3D yang tidakoptimal. Menggunakan metode penelitian kuantitatif dengan pendekataneksperimental. Culling Occlusion adalah teknik yang menyembunyikan objekyang tidak terlihat oleh pemain, sementara LOD adalah teknik yangmenampilkan versi objek dengan detail yang berkurang berdasarkan jarakantara pemain dan objek.Hasil menunjukkan bahwa dengan menggunakanteknik LOD berhasil meningkatkan FPS sebesar 107%,sementara denganteknik Occlusion Culling hanya meningkatkan secara marginal yaitu19%.Temuan ini mengindikasikan bahwa meskipun kedua teknik efektifdalam mengoptimalkan performa, teknik LOD memberikan peningkatan yanglebih signifikan
User Experience Evaluation of YouTube Website Using Eye Tracking Method
Full text linkYouTube is one of the most popular social media in Indonesia, with one of its features being the Clip Feature, which allows users to share 5-60 seconds video snippets, but many users still experience difficulty in accessing this feature. Based on a survey of more than 130 respondents, 60% were unaware of the Clip Feature, 85% had never used it, and 75% had difficulty finding its location in the YouTube interface. This research aims to evaluate the user experience in accessing the Clip Feature on the YouTube website using the Eye Tracking method, as well as analyzing user attention patterns. Through the RealEye.io tool, the results show that the quality of the test data is very good, with an average E-T data integrity value of 90.33% and gaze on screen of 89.73%. Heatmaps and gaze plot analysis show that respondents\u27 attention patterns tend to show confusion, especially when looking for the Clip feature. This is supported by the results of the attention & emotion graphs analysis, which overall show that the average attention level of respondents is at 0.318, with an increase in the emotion of surprise experienced by respondents more than the emotion of happy. Although the Clip Feature offers significant benefits, users still experience difficulties in accessing it, which results in a decreased user experience. This research is expected to provide new recommendations in improving the user experience of YouTube website, specifically to make the Clip feature more accessible and effective to use
Awareness and Responsibility of Food Waste Behavior among College Students
Full text linkIndonesia is facing a substantial issue of food waste. It contributes to environmental problems and loosens up the economy of the country. This study aimed to investigate the food waste behavior of college students as part of young consumers, because consumer behavior established during youth could influence their behavior in the later stages of life. Based on the Norm Activation Model, the research framework was evaluated using structural equation modelling. The result revealed that college students\u27 awareness of consequences and their ascription of responsibility for food waste had a significant impact on their personal norm, which in turn influenced their intention to prevent food waste. The findings are useful to develop strategies to minimize food waste, particularly among student in higher education institution
Automatic Fish Feeding and Temperature Control System for Aquariums Based on Internet of Things (IoT)
Full text linkKeeping fish in aquariums has become one of the people\u27s hobbies. An important factor in fish maintenance is the process of feeding and controlling the temperature of the aquarium. However, with various activities, fish care is often not carried out properly. This study develops an automatic system for feeding and controlling the temperature of the aquarium with goldfish as the test object. This study designs an automatic system to control the temperature and feeding in the aquarium using hardware such as a DS18B20 temperature sensor, load cell, and ultrasonic sensor. This system is controlled by ESP32 for reading sensor data and Arduino Uno for controlling the relay, cooling system, heater, and servo motor. ESP32 reads sensor data and sends it via MQTT to Node-red. Based on this data, the system regulates the temperature by activating the cooler (peltier and water pump) if the temperature is >28℃ and turning off the cooler when the temperature is <26℃. The heater is active if the temperature is <24℃ and stops when the temperature reaches 26℃. Feeding is carried out according to schedule, with servo 1 dropping feed into the load cell until the weight reaches the target weight. After that, servo 2 moves the feed into the aquarium. If the weight has not reached the target, servo 1 continues to be active. Based on the test, the average percentage of error in the temperature sensor is 0,08%, the weight sensor is 1.10%, and the ultrasonic sensor is 1.61%. This system successfully performs four times a day feeding and controls the temperature within the optimal range for goldfish, which is 24-28℃. The test results show that this system functions well and is in accordance with the research objectives.Keeping fish in aquariums has become one of the people\u27s hobbies. An important factor in fish maintenance is the process of feeding and controlling the temperature of the aquarium. However, with various activities, fish care is often not carried out properly. This study develops an automatic system for feeding and controlling the temperature of the aquarium with goldfish as the test object. This study designs an automatic system to control the temperature and feeding in the aquarium using hardware such as a DS18B20 temperature sensor, load cell, and ultrasonic sensor. This system is controlled by ESP32 for reading sensor data and Arduino Uno for controlling the relay, cooling system, heater, and servo motor. ESP32 reads sensor data and sends it via MQTT to Node-red. Based on this data, the system regulates the temperature by activating the cooler (peltier and water pump) if the temperature is >28℃ and turning off the cooler when the temperature is <26℃. The heater is active if the temperature is <24℃ and stops when the temperature reaches 26℃. Feeding is carried out according to schedule, with servo 1 dropping feed into the load cell until the weight reaches the target weight. After that, servo 2 moves the feed into the aquarium. If the weight has not reached the target, servo 1 continues to be active. Based on the test, the average percentage of error in the temperature sensor is 0,08%, the weight sensor is 1.10%, and the ultrasonic sensor is 1.61%. This system successfully performs four times a day feeding and controls the temperature within the optimal range for goldfish, which is 24-28℃. The test results show that this system functions well and is in accordance with the research objectives
Sistem Penyortir Selang Infus Secara Otomatis Berdasarkan Standar Produk 20G dan 25G
No full textPenelitian ini bertujuan untuk merancang dan mengimplementasikan sistem penyortiran salah satu produk medis yaitu selang infus secara otomatis. Sistem yang dikembangkan menggunakan PLC, pneumatik dan Digital Flow Switch. Proses pengujian sistem menggunakan dua standar produk selang infus yaitu standar produk 20G dan 25G. Proses pendeteksian selang infus dikategorikan dalam dua kategori yaitu reject atau accept dengan cara mengamati nilai keluaran pada digital flow switch. Hasil penelitian menunjukkan bahwa produk dikategorikan sebagai produk reject apabila nilai keluaran pada standar produk 20G dan 25G adalah 0,00 L/min. Sementara produk dikategorikan sebagai produk accept apabila nilai keluaran pada standar produk 20G ada pada rentang 4,05 L/min sampai 4,40 L/min sedangkan pada standar produk 25G pada rentang 0,67 L/min sampai 0,92 L/min
Sentiment Analysis on BRImo Application Reviews Using IndoBERT
No full textThe advancement of information technology has significantly impacted various sectors, including digital banking. BRImo, a mobile banking application from Bank Rakyat Indonesia (BRI), has been widely used, generating numerous user reviews that reflect their experiences. This study applies IndoBERT, a transformer-based model specifically designed for the Indonesian language, to analyze sentiment in BRImo user reviews. IndoBERT excels in handling the unique characteristics of the Indonesian language, such as informal and mixed-language usage. The dataset was collected from Kaggle and processed through labeling, data balancing, and splitting into 80% training, 10% validation, and 10% testing data. The IndoBERT model was evaluated using a confusion matrix and achieved 90% accuracy, with F1-scores of 0.89 for negative, 0.91 for neutral, and 0.90 for positive sentiments. Sentiment analysis results indicate that a significant portion of negative reviews highlight issues related to login difficulties, transaction failures, and slow customer service response times. These insights can help BRI enhance application reliability and customer support efficiency. This study demonstrates that IndoBERT is effective in sentiment analysis for Indonesian text and can be utilized to enhance BRImo services by providing deeper insights into user feedback.The advancement of information technology has significantly impacted various sectors, including digital banking. BRImo, a mobile banking application from Bank Rakyat Indonesia (BRI), has been widely used, generating numerous user reviews that reflect their experiences. This study applies IndoBERT, a transformer-based model specifically designed for the Indonesian language, to analyze sentiment in BRImo user reviews. IndoBERT excels in handling the unique characteristics of the Indonesian language, such as informal and mixed-language usage. The dataset was collected from Kaggle and processed through labeling, data balancing, and splitting into 80% training, 10% validation, and 10% testing data. The IndoBERT model was evaluated using a confusion matrix and achieved 90% accuracy, with F1-scores of 0.89 for negative, 0.91 for neutral, and 0.90 for positive sentiments. Sentiment analysis results indicate that a significant portion of negative reviews highlight issues related to login difficulties, transaction failures, and slow customer service response times. These insights can help BRI enhance application reliability and customer support efficiency. This study demonstrates that IndoBERT is effective in sentiment analysis for Indonesian text and can be utilized to enhance BRImo services by providing deeper insights into user feedback
Implementation of Deep Learning with Multilayer Perceptron (MLP) for Heart Disease Prediction Using the SMOTE-ENN Technique
No full textHeart disease is a leading cause of global mortality, with its prevalence increasing annually. This study aims to develop a heart disease prediction model using a Multilayer Perceptron (MLP) combined with the SMOTE-ENN resampling technique to address data imbalance issues. The dataset used was obtained from the UCI Machine Learning Repository and includes patients\u27 clinical and demographic features. The initial dataset consisted of [number of data] records, with an imbalanced class distribution between patients with and without heart disease. After applying SMOTE-ENN, the class distribution became more balanced, allowing the model to learn patterns more effectively. The MLP model was designed with two hidden layers comprising 64 and 32 neurons, respectively, using the ReLU activation function in the hidden layers and a sigmoid function in the output layer. Evaluation results showed that the model achieved an accuracy of 89.47%, precision of 77.78%, recall of 100%, and an F1-score of 87.5%. To validate the effectiveness of SMOTE-ENN, comparisons were made with other methods such as SMOTE and undersampling, as well as baseline models like Logistic Regression and Decision Tree. The results demonstrate that SMOTE-ENN outperforms other techniques in handling class imbalance, leading to better overall model performance.Heart disease is a leading cause of global mortality, with its prevalence increasing annually. This study aims to develop a heart disease prediction model using a Multilayer Perceptron (MLP) combined with the SMOTE-ENN resampling technique to address data imbalance issues. The dataset used was obtained from the UCI Machine Learning Repository and includes patients\u27 clinical and demographic features. The initial dataset consisted of [number of data] records, with an imbalanced class distribution between patients with and without heart disease. After applying SMOTE-ENN, the class distribution became more balanced, allowing the model to learn patterns more effectively. The MLP model was designed with two hidden layers comprising 64 and 32 neurons, respectively, using the ReLU activation function in the hidden layers and a sigmoid function in the output layer. Evaluation results showed that the model achieved an accuracy of 89.47%, precision of 77.78%, recall of 100%, and an F1-score of 87.5%. To validate the effectiveness of SMOTE-ENN, comparisons were made with other methods such as SMOTE and undersampling, as well as baseline models like Logistic Regression and Decision Tree. The results demonstrate that SMOTE-ENN outperforms other techniques in handling class imbalance, leading to better overall model performance
Integrating the CNN Model with the Web for Indonesian Sign Language (BISINDO) Recognition
No full textEffective communication is challenging for deaf individuals in Indonesia, most of whom use Indonesian Sign Language (BISINDO). Sign Language Recognition (SLR) can bridge this communication gap. While Convolutional Neural Networks (CNNs) show high potential for SLR, their practical accessibility remains limited. This research aims to develop a CNN architecture for recognizing BISINDO alphabet signs from static images (still images) and integrate it into an accessible web platform. Using a static vision-based approach, a CNN model was trained on a public dataset (312 images, 26 classes) following standard pre-processing including data augmentation. The model was subsequently integrated into a web interface using Python and the Gradio library. Results demonstrated strong model performance, with validation accuracy reaching 97.44% and a macro-average F1-score of approximately 97.12%. However, classification challenges were identified for visually similar signs (\u27M\u27 and \u27N\u27). The resulting integrated web application proved functional, exhibited low prediction latency, and showed cross-platform compatibility. This study successfully demonstrates the development of an accurate DL model for static BISINDO alphabet recognition and its practical implementation via a web platform. This contributes to reducing the accessibility gap in SLR technology. Future research is recommended to utilize larger, more varied datasets and explore dynamic sign recognition
A Random Forest-Based Predictive Model for Student Academic Performance: A Case Study in Indonesian Public High Schools
No full textThe rapid advancement of information technology has transformed education by providing tools to accurately predict students\u27 academic performance. This study aims to develop a system for predicting academic achievement using the Random Forest algorithm, with a case study at SMAN 1 Aceh Barat Daya and SMAN 3 Aceh Barat Daya. Data from 632 student report cards for grades X and XI in the second semester of the 2023/2024 academic year were used, covering subjects such as Mathematics, Indonesian Language, and others, divided into 80% training data (506 samples) and 20% test data (136 samples). The research methodology involved data preprocessing, training the Random Forest model using entropy and information gain to construct decision trees, and performance evaluation using metrics such as accuracy, precision, and recall. The implementation resulted in a web-based application using Python and Flask, featuring an interactive interface and decision tree visualization. Testing on 136 test samples achieved an accuracy of 87.40%, with 111 correct predictions, 16 false positives, and 0 false negatives, demonstrating the model\u27s reliability in identifying high-achieving students without missing potential. This research is expected to assist schools in identifying outstanding students, making data-driven decisions, and designing more effective educational strategies.The rapid advancement of information technology has transformed education by providing tools to accurately predict students\u27 academic performance. This study aims to develop a system for predicting academic achievement using the Random Forest algorithm, with a case study at SMAN 1 Aceh Barat Daya and SMAN 3 Aceh Barat Daya. Data from 632 student report cards for grades X and XI in the second semester of the 2023/2024 academic year were used, covering subjects such as Mathematics, Indonesian Language, and others, divided into 80% training data (506 samples) and 20% test data (136 samples). The research methodology involved data preprocessing, training the Random Forest model using entropy and information gain to construct decision trees, and performance evaluation using metrics such as accuracy, precision, and recall. The implementation resulted in a web-based application using Python and Flask, featuring an interactive interface and decision tree visualization. Testing on 136 test samples achieved an accuracy of 87.40%, with 111 correct predictions, 16 false positives, and 0 false negatives, demonstrating the model\u27s reliability in identifying high-achieving students without missing potential. This research is expected to assist schools in identifying outstanding students, making data-driven decisions, and designing more effective educational strategies