Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Comparative Study of Linear Regression, SVR, and XGBoost for Stock Price Prediction After a Stock Split
Get PDFThis study aims to identify the most effective regression method for predicting the closing stock price of Bank Central Asia (BBCA) following the stock split event on October 12, 2021. Accurate post-split price predictions are crucial for helping investors comprehend new market dynamics, yet there is limited research evaluating the performance of regression models on BBCA’s stock after such corporate actions. Using data obtained through web scraping from the Indonesia Stock Exchange, this study tested three regression algorithms Linear Regression, Support Vector Regression, and XGBoost Regressor on post-split data. The selected input features were open_price, first_trade, high, low, and volume, while the target was close_price. The dataset was divided using an 80:20 train-test split and evaluated with RMSE, MAPE, and R-squared metrics. Results showed that Linear Regression achieved the best performance RMSE: 50.41, MAPE: 0.0048, R²: 0.9971, followed by XGBoost RMSE: 69.12, MAPE: 0.0058, R²: 0.9946, and SVR RMSE: 80.98, MAPE: 0.0069, R²: 0.9925. These findings indicate that BBCA’s post-split stock data exhibits a linear pattern, making Linear Regression the most suitable and efficient method. This suggests that simpler models can outperform more complex algorithms when applied to stable and structured financial datasets.This study aims to identify the most effective regression method for predicting the closing stock price of Bank Central Asia (BBCA) following the stock split event on October 12, 2021. Accurate post-split price predictions are crucial for helping investors comprehend new market dynamics, yet there is limited research evaluating the performance of regression models on BBCA’s stock after such corporate actions. Using data obtained through web scraping from the Indonesia Stock Exchange, this study tested three regression algorithms Linear Regression, Support Vector Regression, and XGBoost Regressor on post-split data. The selected input features were open_price, first_trade, high, low, and volume, while the target was close_price. The dataset was divided using an 80:20 train-test split and evaluated with RMSE, MAPE, and R-squared metrics. Results showed that Linear Regression achieved the best performance RMSE: 50.41, MAPE: 0.0048, R²: 0.9971, followed by XGBoost RMSE: 69.12, MAPE: 0.0058, R²: 0.9946, and SVR RMSE: 80.98, MAPE: 0.0069, R²: 0.9925. These findings indicate that BBCA’s post-split stock data exhibits a linear pattern, making Linear Regression the most suitable and efficient method. This suggests that simpler models can outperform more complex algorithms when applied to stable and structured financial datasets
Green Supply Chain Management (GSCM) Practices on Sustainability Performance in the Manufacturing Industry of Batam City
Get PDFThe research aims to explore the effect of seven dimensions Green Supply Chain Management (GSCM) practices on three dimensions of sustainability performance, namely environmental, economic, and social. The seven dimensions of GSCM practices include green purchasing, green manufacturing, green marketing, green distribution and packaging, internal environmental management, environmental education, and investment recovery. The research sample was sourced from 70 manufacturing companies categorized as large and medium industries in Batam City. Data collection was done by distributing questionnaires. Data were statistically evaluated by applying Structural Equation Modeling-Partial Least Squares (SEM-PLS) methodology employing SmartPLS 4. Findings from the study indicate that every dimension of GSCM practice positively and significantly impacts on at least one dimension of sustainability performance. From seven dimensions of GSCM practices tested, green marketing and green distribution & packaging positively and significantly affect all three dimensions of sustainability performance
Pemberdayaan Masyarakat di Kawasan Agropolitan Kabupaten Kediri
Get PDFKediri is one of the regencies in East Java Province with a variety of potential natural resources. In Kediri Regency, there are various potentials and natural resources that can support the welfare of the community spread across several sub-districts. The potential and natural resources that exist can be used as an agropolitan area in order to empower the community. Through the implementation method based on regional area design, sector implementation, and impact on the community which are the three main phases in an effort to increase the spirit of skilled and creative work for the community in the agropolitan area of Kediri Regency. The potential and natural resources that can be maximized for export, processing, and creation into productivity in the agropolitan area include: 1) Pineapple, Papaya, and Turmeric which can be maximized in the sub-districts of Ngancar, Wates, Plosoklaten, and Kandat, 2) Oyster Mushrooms, Corn, and Sugar Cane which can be maximized in the sub-districts of Ngadiluwih, Ringinrejo, and Kras, 3) Chili and Vegetables which can be maximized in the sub-districts of Pare, Kandangan, Puncu, and Kepung, 4) Rice and Secondary Crops which can be maximized in the sub-districts of Pagu, Plemahan, Papar, and Purwoasri, and 5) Podang Mango, Coffee, Orange, and Cassava which can be maximized in the sub-districts of Semen, Grogol, Banyakan, Tarokan, and Mojo. Through the distribution map that has been designed by the Kediri Regency Government, it at least provides an illustration to be maximized massively and sustainably in order to increase income and the economy for the people of Kediri Regency.Kediri merupakan salah satu Kabupaten di Provinsi Jawa Timur dengan beragam hasil potensi sumber daya alam. Di Kabupaten Kediri terdapat berbagai potensi dan sumber daya alam yang dapat menunjang kesejahteraan masyarakat yang tersebar di beberapa kecamatan. Potensi dan sumber daya alam yang ada dapat dijadikan sebagai kawasan agropolitan agar dapat memberdayakan masyarakat. Melalui metode pelaksanaan berbasis desain kawasan wilayah, implementasi sektor, dan dampak bagi masyarakat yang merupakan tiga fase utama dalam upaya meningkatkan semangat kerja secara terampil dan kreatif bagi masyarakat di kawasan agropolitan Kabupaten Kediri. Potensi dan sumber daya alam yang dapat dimaksimalkan untuk bisa diekspor, diolah, dan dikreasi menjadi produktivitas di kawasan agropolitan diantaranya ada: 1) Nanas, Pepaya, dan Kunyit yang dapat dimaksimalkan di kecamatan Ngancar, Wates, Plosoklaten, dan Kandat, 2) Jamur Tiram, Jagung, dan Tebu yang dapat dimaksimalkan di kecamatan Ngadiluwih, Ringinrejo, dan Kras, 3) Cabai dan Sayur Mayur yang dapat dimaksimalkan di kecamatan Pare, Kandangan, Puncu, dan Kepung, 4) Padi dan Palawija yang dapat dimaksimalkan di kecamatan Pagu, Plemahan, Papar, dan Purwoasri, dan 5) Mangga Podang, Kopi, Jeruk, dan Ubi Kayu yang dapat dimaksimalkan di kecamatan Semen, Grogol, Banyakan, Tarokan, dan Mojo. Melalui peta persebaran yang telah di desain oleh Pemerintah Kabupaten Kediri setidaknya memberikan ilustrasi utnuk dapat dimaksimalkan secara masif dan berkelanjutan agar dapat meningkatkan pendapatan dan perekonomian bagi masyarakat Kabupaten Kediri. 
PEMBUATAN DAN PENGUJIAN VARIASI PEMBEBANAN TERHADAP WAKTU YANG DIBUTUHKAN MESIN PEMISAH SARI PATI KEDELAI UNTUK PABRIK TAHU
Get PDFIn Lohbener Village, Indramayu Regency, small and medium enterprises (SMEs) produce tofu using traditional methods, where the soybean pulp filtering process requires significant time and physical effort. To address this issue, a soybean pulp filtering machine was developed to improve filtering efficiency. The machine was tested with three different input loads—5 kg, 10 kg, and 15 kg—at a constant speed of 57 rpm. The filtering times recorded were 3 minutes 10 seconds, 5 minutes 17 seconds, and 8 minutes 21 seconds, respectively, determined by the point at which no more liquid discharged from the system. The final weights of the filtered soybean pulp were 1.5 kg, 5.7 kg, and 7.4 kg, respectively. These weight differences were influenced by an unbalanced soybean-to-water ratio, which affected extraction effectiveness. The average moisture content of the filtered pulp was 18% for the 5 kg load, 14.90% for 10 kg, and 15.17% for 15 kg. These results indicate that the use of the filtering machine significantly improves filtering efficiency compared to manual methods, in terms of time, output, and pulp quality. Overall, the machine has a positive impact on increasing productivity and product quality in tofu production at the SME level.Indramayu desa Lohbener terdapat umkm pembuatan tahu dimana produksi tahu ini menggunakan proses tradisional yang terbilang pada saat proses produksi membutuhkan waktu yang lama dan berat bagi para pekerjanya terutama pada proses penyaringan. Masalah pada prsoses ini dapat diatasi dengan pembuatan mesin pengayakan sari pati kedelai. Dari hasil proses pembuatan mesin dengan pengujian variasi pembebanan terhadap beban 5, 10, 15 kg di kecepatan yang sama, dapat disimpulkan waktu yang efisien pada saat proses pengayakan ada di 3 menit 10 detik, 5 menit 17 detik, 8 menit 21 detik yang dimana tidak ada lagi air yang mengalir pada saat proses pengayakan. Hasil perbandingan berat sari pati kedelai setelah dilakukan pengayakan dengan kecepatan pengayakan 57 Rpm di dapatkan, pada pengujian 1 dengan berat awal sari pati kedelai sebesar 5 Kg menjadi 1,5 Kg, sedangkan pada pengujian 2 dengan berat awal sari pati kedelai sebesar 10 Kg menjadi 5,7 Kg dan pada pengujian 3 dengan berat awal sari pati kedelai sebesar 15 Kg menjadi 7,4 Kg. Perbedaan berat yang cukup signifikan di pengaruhi oleh campuran antara kedelai dengan air yang tidak seimbang sehingga mempengaruhi berat hasil proses pengayakan. Maka dari data tersebut menunjukkan bahwa pada proses penyaringan menggunakan mesin penyaring dapat tersaring dengan baik dibandingkan dengan penyaringan manual dan dapat dilihat presentase keefektifan dari penyaringan tersebut. Hasil dari rata-rata kadar air variasi beban 5 Kg dikisaran 18%, untuk rata-rata kadar air variasi 10 Kg dikisaran 14,90%, dan untuk rata-rata kadar air variasi 15 Kg dikisaran 15,17%
Improving Efficient Ship Detection Performance Using Contextual Transformers for Maritime Surveillance
No full textShip surveillance plays a crucial role in enhancing defense systems in coastal areas. An automatic vessel detection system is necessary to accurately identify vessels and their categories, typically utilizing a reliable computer vision system. The nano version of YOLO11 has emerged as one of the object detection methods that officially provides lightweight computing, but still has limitations in extracting complex features. Contextual Transformer (CoT) efficiently utilizes long-range relationships, thereby enhancing feature discrimination performance. This study proposes a vessel detection system by modifying the YOLO11 architecture using the Contextual Transformer block. This work introduces YOLO11-Pico, a lighter version of nano, with channel size adjustments at certain stages for further efficiency. The proposed CoT block applies fewer multiplication mapping operations, which are capable of representing global features to obtain richer contextual information. The SeaShips dataset is used as the source of data for model training and evaluation. Experimental results demonstrate that the proposed model YOLO11-pico-CoT achieves superior performance compared to prominent lightweight YOLO architectures, including the YOLO11n baseline, YOLOv5n, YOLOv10n, and the latest YOLOv12n. The integration of CoT contributes positively to improving the accuracy of ship category and location predictions, achieving 0.964 mAP50 and 0.714 mAP50:95. Additionally, efficiency evaluations show that the proposed module is computationally lighter and has fewer parameters, specifically 1,711,250 parameters while operating at 3.97 FPS, giving it an advantage in terms of capabilities over the comparison methods.Ship surveillance plays a crucial role in enhancing defense systems in coastal areas. An automatic vessel detection system is necessary to accurately identify vessels and their categories, typically utilizing a reliable computer vision system. The nano version of YOLO11 has emerged as one of the object detection methods that officially provides lightweight computing, but still has limitations in extracting complex features. Contextual Transformer (CoT) efficiently utilizes long-range relationships, thereby enhancing feature discrimination performance. This study proposes a vessel detection system by modifying the YOLO11 architecture using the Contextual Transformer block. This work introduces YOLO11-Pico, a lighter version of nano, with channel size adjustments at certain stages for further efficiency. The proposed CoT block applies fewer multiplication mapping operations, which are capable of representing global features to obtain richer contextual information. The SeaShips dataset is used as the source of data for model training and evaluation. Experimental results demonstrate that the proposed model YOLO11-pico-CoT achieves superior performance compared to prominent lightweight YOLO architectures, including the YOLO11n baseline, YOLOv5n, YOLOv10n, and the latest YOLOv12n. The integration of CoT contributes positively to improving the accuracy of ship category and location predictions, achieving 0.964 mAP50 and 0.714 mAP50:95. Additionally, efficiency evaluations show that the proposed module is computationally lighter and has fewer parameters, specifically 1,711,250 parameters while operating at 3.97 FPS, giving it an advantage in terms of capabilities over the comparison methods
Optimization of Rice Field Irrigation Based on Fuzzy Logic and the Internet of Things Through Water Level Analysis
No full textThe low efficiency of conventional irrigation systems often results in water waste and decreased rice productivity. The research was carried out by designing an automatic monitoring and control system using a water level sensor, a Raspberry Pi Pico W microcontroller, a water pump, and a Blynk application as a real-time monitoring medium. Water level data is processed by fuzzy logic method to categorize low, normal, or high conditions, so that the system can adjust the water pump adaptively according to the needs of the land. The results of the study show that the integration of IoT and fuzzy logic is able to improve water use efficiency, maintain soil moisture at optimal conditions, and support better rice growth. The system has also been proven to be accurate in the classification of water conditions with a success rate above 90%. Thus, this research contributes to the development of smart agricultural technologies that can increase productivity while supporting sustainable agricultural practices
A Comparative Analysis of Character and Word-Based Tokenization for Kawi-Indonesian Neural Machine Translation
No full textPreserving regional languages is a strategic step in preserving cultural heritage while expanding access to knowledge across generations. One approach that can support this effort is the application of automatic translation technology to digitize and learn local language texts. This study compares two tokenization strategies, word-based and character-based on a Kawi–Indonesian translation model using the FLAN-T5-Small Transformer architecture. The dataset used consists of 4,987 preprocessed sentence pairs, trained for 10 epochs with a batch size of 8. Statistical analysis shows that Kawi texts have an average length of 39.6 characters (5.4 words) per sentence, while Indonesian texts have an average length of 54.9 characters (7.5 words). These findings suggest that Kawi sentences tend to be lexically dense, with low word repetition and high morphological variation, which can increase the learning complexity of the model. Evaluation using BLEU and METEOR metrics shows that the model with word-based tokenization achieved a BLEU score of 0.45 and a METEOR score of 0.05, while the character-based model achieved a BLEU score of 0.24 and a METEOR score of 0.04. Although the dataset size has increased compared to previous studies, these results indicate that the additional data is not sufficient to overcome the limitations of the semantic representation of the Kawi language. Therefore, this study serves as an initial baseline that can be further developed through subword tokenization approaches, dataset expansion, and training strategy optimization to improve the quality of local language translations in the future
Rancang bangun pendeteksi suhu dan gas amonia dengan metode fuzzy logic pada kandang ayam broiler berbasis iot
No full textTemperature, humidity, and ammonia gas are critical factors that affect the health and productivity of broiler chickens. Manual monitoring commonly carried out by farmers is inefficient and prone to errors, thus requiring an automatic system capable of real-time detection. This study developed a monitoring system for temperature, humidity, and ammonia levels in broiler chicken coops based on the IoT using the Sugeno fuzzy logic method. The system was designed with a NodeMCU ESP8266 microcontroller integrated with a DHT22 sensor for temperature and humidity, and a MQ-135 sensor for ammonia gas. Measurement data are displayed in real time through an LCD, Blynk application, and ThingSpeak allowing remote monitoring. The decision-making process is carried out using Sugeno fuzzy logic to classify coop conditions into safe, alert, or danger. Experimental results show MAPE values of 2,7% for temperature and 3,5% for humidity. The system operated continuously for 14 days and generated 1,265 consistent monitoring data points. Thus, the proposed system assists in automatic coop monitoring and supports increased broiler chicken productivity
Implementasi Sistem Pemantauan Energi Listrik dan Perbaikan Faktor Daya Berbasis Android dengan MIT App Inventor dan Firebase
No full textAs inductive loads like washing machines and water pumps become more common, they utilize more reactive energy and lower the power factor. If the power factor is less than 0,85 PLN, electrical protection devices like MCBs and ELCBs could stop working. To utilize less power, things need to change. Adding a capacitor bank to add reactive power can make the power factor value better. Using the MIT App Inventor and Firebase platforms, this study built an app that lets you keep track of and improve your power factor. The app had an average data latency of 11,20 seconds. After 11 load changes, the average power factor went up from 0,64 (before improvement) to 0,88 (after). So, this method for optimizing power factor helps to get a low power factor up to the right level.Listrik merupakan salah satu kebutuhan pokok dalam kehidupan. Penggunaan beban induktif seperti mesin cuci, pompa air, kipas angin, dan lemari es menyebabkan peningkatan kebutuhan energi reaktif serta penurunan faktor daya. Pada sistem rumah tangga, faktor daya yang rendah dapat memicu perangkat proteksi listrik seperti MCB dan ELCB terjadi trip. PLN menetapkan nilai minimum faktor daya sebesar >0,85. Tujuan utama dari perbaikan faktor daya adalah untuk meningkatkan kapasitas sistem dan mengoptimalkan konsumsi daya listrik. Salah satu solusi yang dapat diterapkan adalah pemasangan bank kapasitor untuk menyuplai daya reaktif ke dalam sistem, sehingga nilai faktor daya dapat meningkat. Pada penelitian ini, MIT App Inventor dan Firebase digunakan sebagai aplikasi pemantauan. Proses pengiriman data ke aplikasi Android menunjukkan rata-rata waktu tunda sebesar 11,20 detik. Berdasarkan pengujian dengan 11 variasi beban, diperoleh rata-rata nilai faktor daya yang lebih baik setelah dilakukan perbaikan
Public Opinion on The MBG Program: Comparative Evaluation of InSet and VADER Lexicon Labeling Using SVM on Platform X
No full textThis study aims to examine public opinion regarding the MBG program on platform X by utilizing the Support Vector Machine (SVM) algorithm using two sentiment labeling methods, namely InSet Lexicon and VADER Lexicon. The data was then divided into 70% for training and 30% for testing, and extracted using Term Frequency–Inverse Document Frequency (TF-IDF) to convert the text into numerical representations. The SVM model was trained on both labeled data sets to compare their performance based on evaluation metrics such as accuracy, precision, recall, and F1 score. The results show that labeling with VADER produces a more dominant number of neutral sentiments, while InSet Lexicon produces a more balanced distribution between positive, negative, and neutral sentiments. At the modeling stage, SVM with InSet labels achieved an accuracy of 80.10%, with precision of 0.81, recall of 0.80, and an F1 score of 0.79. Meanwhile, SVM with VADER labels achieved an accuracy of 93.83%, precision of 0.94, recall of 0.94, and an F1 score of 0.93. Although VADER showed higher accuracy values, InSet Lexicon is considered more efficient and relevant for sentiment analysis in Indonesia because it is capable of producing more balanced and contextual classifications.This study aims to examine public opinion regarding the MBG program on platform X by utilizing the Support Vector Machine (SVM) algorithm using two sentiment labeling methods, namely InSet Lexicon and VADER Lexicon. The data was then divided into 70% for training and 30% for testing, and extracted using Term Frequency–Inverse Document Frequency (TF-IDF) to convert the text into numerical representations. The SVM model was trained on both labeled data sets to compare their performance based on evaluation metrics such as accuracy, precision, recall, and F1 score. The results show that labeling with VADER produces a more dominant number of neutral sentiments, while InSet Lexicon produces a more balanced distribution between positive, negative, and neutral sentiments. At the modeling stage, SVM with InSet labels achieved an accuracy of 80.10%, with precision of 0.81, recall of 0.80, and an F1 score of 0.79. Meanwhile, SVM with VADER labels achieved an accuracy of 93.83%, precision of 0.94, recall of 0.94, and an F1 score of 0.93. Although VADER showed higher accuracy values, InSet Lexicon is considered more efficient and relevant for sentiment analysis in Indonesia because it is capable of producing more balanced and contextual classifications