Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Data Streaming Pipeline Model Using DBSTREAM-Based Online Machine Learning for E-Commerce User Segmentation
No full textThe rapid development of information technology has driven major transformations in the digital business sector, particularly e-commerce. Consumers who shop at e-commerce sites generally have different characteristics, behaviors, and needs. Analyzing the behavior of each consumer is difficult to do manually, requiring an automation system that can help identify consumer behavior patterns adaptively. However, most customer segmentation approaches still rely on batch learning methods based on static data, making them unable to quickly adapt to changes in user behavior. This study aims to design a streaming data pipeline based on Online Machine Learning (OML) integrated with the Density-Based Clustering for Data Streams (DBSTREAM) algorithm to produce adaptive e-commerce user segmentation. The system was developed using Python with RabbitMQ as a real-time data stream simulator, MongoDB for storing results, and Streamlit as a visualization interface. The clustering process was performed incrementally using DBSTREAM, then stabilized through Hierarchical Agglomerative Clustering (HAC) to avoid over-segmentation. Evaluation using the Silhouette Coefficient and Davies-Bouldin Index (DBI) shows that the optimal model for the cluster threshold is in the range of 0.6 to 0.8 and for the fading factor is 0.0005 or even smaller, such as 0.0003. The evaluation results obtained a Silhouette value of -0.1125 and a DBI of 0.2796. These results prove that DBSTREAM-based OML integration is capable of forming consumer behavior segmentation efficiently and adaptively to continuous and real-time changes in streaming data
Comparative Analysis of EfficientNet-B0 and ViT-B16 for Multiclass Classification of Green Coffee Beans
No full textGreen coffee bean classification plays an important role in the coffee supply chain, as bean quality has a direct impact on the taste and final quality of the product. The USK-Coffee dataset, which consists of four bean object classes defect, longberry, peaberry, and premium, is photographed under varied lighting conditions and capture angles, thus challenging the accuracy of conventional visual models. Although lightweight CNN models have been used, not many studies have directly compared transformer-based architectures (ViT-B16) and modern efficient CNNs (EfficientNet-B0) for green coffee bean classification under real conditions. With transfer learning strategy, image augmentation (resize, flip, rotation, color jitter, random crop), and normalization, we evaluate the performance of both models on the dataset. ViT-B16 achieved 85% accuracy on the test data (F1-score 0.85), with a fast batch inference latency of 0.0074 seconds per batch. EfficientNet-B0 achieved 87% accuracy (F1-score 0.87), with a slower batch latency (0.0106 seconds per batch). However, EfficientNet-B0 is significantly faster for single image inference (real-time) (0.035 seconds) compared to ViT-B16 (0.426 seconds). This trade-off higher accuracy/faster single inference on EfficientNet-B0 vs. faster batch processing on ViT-B16 shows that both are feasible for edge computing-based classification systems.Green coffee bean classification plays an important role in the coffee supply chain, as bean quality has a direct impact on the taste and final quality of the product. The USK-Coffee dataset, which consists of four bean object classes defect, longberry, peaberry, and premium, is photographed under varied lighting conditions and capture angles, thus challenging the accuracy of conventional visual models. Although lightweight CNN models have been used, not many studies have directly compared transformer-based architectures (ViT-B16) and modern efficient CNNs (EfficientNet-B0) for green coffee bean classification under real conditions. With transfer learning strategy, image augmentation (resize, flip, rotation, color jitter, random crop), and normalization, we evaluate the performance of both models on the dataset. ViT-B16 achieved 85% accuracy on the test data (F1-score 0.85), with a fast batch inference latency of 0.0074 seconds per batch. EfficientNet-B0 achieved 87% accuracy (F1-score 0.87), with a slower batch latency (0.0106 seconds per batch). However, EfficientNet-B0 is significantly faster for single image inference (real-time) (0.035 seconds) compared to ViT-B16 (0.426 seconds). This trade-off higher accuracy/faster single inference on EfficientNet-B0 vs. faster batch processing on ViT-B16 shows that both are feasible for edge computing-based classification systems
Provincial Tax Performance in Indonesia and Determinants of Tax Revenue
No full textThis research seeks to assess the efficacy of tax revenues across 34 provinces in Indonesia and analyze the determinants affecting provincial tax collections on a national scale. This research utilizes yearly data released by the Central Statistics Agency of Indonesia from 2019 to 2023. The Tax Performance Index assesses tax revenue performance over a five-year period, using panel data regression analysis to evaluate the impact of various factors. A panel data regression analysis was performed using the fixed effects model. The results of the regression analysis indicate that most provinces have not yet achieved optimal tax revenue levels. The regression study findings indicate that Gross Domestic Product (GDP) per capita, Human Development Index, Percentage of Workforce, and Gini Index strongly influence tax revenues, both individually and together. The research findings suggest that the government must consider both non-economic and economic factors that may influence tax collections
Leverage as Moderation on the Effect Firm Size, Managerial Ownership and Conflict of Interest on Accounting Conservatism
No full textThis study examines how firm size, managerial ownership, and conflict of interest influence the application of accounting conservatism, with leverage as a moderating variable. The population studied comprised 125 manufacturing companies in the non-cyclical consumer sector, listed on the Indonesia Stock Exchange (IDX) between 2019 and 2023. Using purposive sampling, 41 companies were selected, yielding 2025 units of analysis. The food and beverage sector was chosen for its stable demand, despite challenges such as strict regulations, fluctuations in raw material prices, and growing health and sustainability awareness. The analytical tools used to test the hypotheses were multiple regression and moderating-variable regression analyses in IBM SPSS 26. The results of the study indicate that firm size, managerial ownership, and conflict of interest do not affect accounting conservatism. Leverage is unable to moderate the relationship between firm size and managerial ownership on accounting conservatism. However, leverage moderated the effect of conflict of interest on accounting conservatism, weakening it. The results of the study indicate that firm size, managerial ownership, and conflict of interest do not affect accounting conservatism. Leverage is unable to moderate the relationship between firm size and managerial ownership on accounting conservatism. However, leverage can moderate the effect of conflict of interest on accounting conservatism, thereby weakening it. The results of this study indicate that a larger size does not guarantee that a company will apply the principle of conservatism. Managerial decisions and internal company policies often have a greater influence than size. Managerial ownership also cannot explain how accounting conservatism is applied, because low managerial ownership makes managers less conservative in preparing financial statements. Company managers currently receive bonuses because of their sense of ownership of the company, not only because of increased profits. A conflict of interest within the company does not always affect accounting conservatism, depending on specific conditions. When a company has low debt, management may feel freer to make more optimistic decisions because they do not face financial pressure from creditors, thereby reducing conflicts of interest
Digitalisasi Pemasaran Produk UMKM: Inovasi, Branding, dan Ekspansi Pasar
No full textMicro, Small, and Medium Enterprises (MSMEs) are a vital sector in Indonesia’s economy; however, they still face significant challenges in effectively marketing their products in the digital era. The main issues encountered include a lack of product innovation, weak branding strategies, and limited market expansion through digital platforms. MSME products often do not follow market trends and lack unique selling points, while the brands built tend to be unappealing and inconsistent both visually and narratively. On the other hand, the utilization of digital technology to reach wider markets remains limited due to low digital literacy and the absence of effective marketing strategies. This program aims to provide a comprehensive solution through training and mentoring in digital marketing, focusing on innovative product development, brand identity strengthening, and the use of social media, e-commerce, and SEO. It is expected that this initiative will enhance MSMEs\u27 competitiveness, expand their market reach, and establish a more adaptive and sustainable business ecosystem amid technological advancementUsaha Mikro, Kecil, dan Menengah (UMKM) merupakan sektor vital dalam perekonomian Indonesia, namun masih menghadapi tantangan besar dalam memasarkan produknya secara efektif di era digital. Permasalahan utama yang dihadapi adalah kurangnya inovasi produk, lemahnya strategi branding, dan keterbatasan ekspansi pasar melalui platform digital. Inovasi produk UMKM sering kali tidak mengikuti tren pasar dan minim pembeda, sementara brand yang dibangun cenderung kurang menarik dan tidak konsisten secara visual maupun naratif. Di sisi lain, pemanfaatan teknologi digital untuk menjangkau pasar yang lebih luas masih sangat terbatas akibat rendahnya literasi digital dan kurangnya strategi pemasaran yang tepat. Program ini bertujuan untuk memberikan solusi menyeluruh melalui pelatihan dan pendampingan digitalisasi pemasaran, dengan fokus pada pengembangan produk inovatif, penguatan identitas merek, serta pemanfaatan media sosial, e-commerce, dan SEO. Diharapkan kegiatan ini dapat meningkatkan daya saing UMKM, memperluas jangkauan pasar, serta membentuk ekosistem usaha yang adaptif dan berkelanjutan di tengah perkembangan teknolog
Implementasi Metode Pose to Pose dalam Perancangan Ekspresi dan Gestur Pada Karakter Deka dan Dara dari Animasi 3D Ficusia
No full textAs technology builds up, the multimedia process is also evolving. Animation is one of the entertainment media that people of all ages enjoy. All of them appreciate animation. One of the issues that occurs in animating is the expression and gestures that aren\u27t strong enough. An animation\u27s message and atmosphere can be conveyed through these two elements. One technique animators can use is called the pose to pose method. The purpose of this paper is to describe the expressions and gestures process of characters from the 3D Animation Series Ficusia using the pose to pose method. Two characters from this animation were used to compose this research by setting the keyframe to identify the key pose, timing, and spacing, and considering the total number of frames used. This research was conducted using Blender.
Perancangan Dan Analisis Desain Maskot Sebagai Representasi Identitas Visual Bagian Promosi Badan Pengusahaan Batam
Get PDFHingga saat ini, kegiatan promosi Badan Pengusahaan Batam (BP Batam) yang dilakukan oleh Bagian Promosi BP Batam belum memiliki maskot resmi yang dapat merepresentasikan identitas visual Promosi BP Batam. Perancangan maskot promosi BP Batam sebagai langkah yang tepat untuk merepresentasikan identitas visual yang kuat dalam meningkatkan investasi dan membangun citra positif instansi kepada masyarakat terutama masyarakat yang bergerak di dunia usaha, baik lokal maupun asing. Maskot Promosi BP Batam terinspirasi dari logo BP Batam yang merupakan seekor burung elang laut yang bernama SI PRIMA sebagai wujud tugas Bagian Promosi BP Batam untuk memberikan pelayanan PRIMA dalam meningkatkan INVESTASI, promosi dan industri kota Batam. Maskot SI PRIMA bergambar burung elang laut jantan dengan memakai baju dinas BP Batam dan dipadupadankan dengan ciri khas adat Melayu. Perancangan maskot menggunakan metode design thinking dan metode kuantitatif deskriptif untuk mengukur tingkat ketercapaian tujuan perancangan dan analisis maskot SI PRIMA sebagai representasi identitas visual Promosi BP Batam. Hasil pengujian user testing pada penelitian ini menunjukkan dari aspek visual dan aspek media bahwa berdasarkan pada skala ketercapaian tujuan perancangan maskot SI PRIMA dalam bentuk 2 dimensi sebagai representasi identitas visual Promosi BP Batam berhasil berada pada tingkat positif atau sangat baik dalam rentang 75,1% - 100%
Game-Based Learning for Mathematics Lesson on 3rd Grade Elementary School
Get PDFIntegrating cutting-edge strategies to improve students\u27 learning experiences has become increasingly important in the ever-changing world of education. This study investigates how third-grade students at SD Pius Purbalingga can benefit from using game-based learning as an instructional strategy to improve their mathematical education. The study focuses on how to hold children\u27 attention and improve their knowledge of mathematics. The main subject of this study is the effectiveness of educational games in enhancing elementary school student\u27s understanding of mathematics. A mathematics game was created to solve this problem by actively involving pupils and reiterating key mathematical ideas. This game-based strategy aimed to create an engaged and enjoyable learning experience for third-grade pupils with acceptable cognitive capacities. The findings suggest that students who played the math game significantly increased their involvement, comprehension, and memorization of mathematical ideas. This study adds to the growing evidence supporting using educational games as useful tools in mathematics instruction. The study\u27s findings revealed increased academic performance among students, with male students experiencing a rise of 2.4% in their overall scores. In contrast, female students demonstrated a significantly higher increase of 8.5%, indicating a more pronounced advancement in their academic performance.Integrating cutting-edge strategies to improve students\u27 learning experiences has become increasingly important in the ever-changing world of education. This study investigates how third-grade students at SD Pius Purbalingga can benefit from using game-based learning as an instructional strategy to improve their mathematical education. The study focuses on how to hold children\u27 attention and improve their knowledge of mathematics. The main subject of this study is the effectiveness of educational games in enhancing elementary school student\u27s understanding of mathematics. A mathematics game was created to solve this problem by actively involving pupils and reiterating key mathematical ideas. This game-based strategy aimed to create an engaged and enjoyable learning experience for third-grade pupils with acceptable cognitive capacities. The findings suggest that students who played the math game significantly increased their involvement, comprehension, and memorization of mathematical ideas. This study adds to the growing evidence supporting using educational games as useful tools in mathematics instruction. The study\u27s findings revealed increased academic performance among students, with male students experiencing a rise of 2.4% in their overall scores. In contrast, female students demonstrated a significantly higher increase of 8.5%, indicating a more pronounced advancement in their academic performance
Evaluation of Telecommunication Customer Churn Classification with SMOTE Using Random Forest and XGBoost Algorithms
Get PDFCompetition in the telecommunications industry, particularly among Internet Service Providers (ISPs), significantly influences customer churn, which negatively impacts revenue, profitability, and business sustainability. An effective approach to mitigate churn involves identifying potential churners early, enabling companies to implement strategic retention measures. However, predicting churn can be challenging due to the limited data available on churned customers. This study aims to predict customers likely to terminate or discontinue their subscriptions, focusing on addressing data imbalance using the Synthetic Minority Over-Sampling Technique (SMOTE). The dataset, sourced from Kaggle, comprises 21 attributes and 7,034 entries. The pre-processing phase includes data cleaning, feature encoding, and the implementation of Random Forest and XGBoost algorithms after data balancing with SMOTE. The findings reveal that the XGBoost algorithm achieves a prediction accuracy of 82%, outperforming Random Forest with 81%. Key factors influencing churn include Contract, TotalCharges, and tenure. The study concludes by emphasizing the significance of contract flexibility and the need to prioritize customers with high total costs or extended subscription periods to reduce churn rates. Future research is encouraged to investigate alternative methods for handling data imbalance and to explore advanced machine learning algorithms to further enhance prediction accuracy and the effectiveness of customer retention strategies.Competition in the telecommunications industry, particularly among Internet Service Providers (ISPs), significantly influences customer churn, which negatively impacts revenue, profitability, and business sustainability. An effective approach to mitigate churn involves identifying potential churners early, enabling companies to implement strategic retention measures. However, predicting churn can be challenging due to the limited data available on churned customers. This study aims to predict customers likely to terminate or discontinue their subscriptions, focusing on addressing data imbalance using the Synthetic Minority Over-Sampling Technique (SMOTE). The dataset, sourced from Kaggle, comprises 21 attributes and 7,034 entries. The pre-processing phase includes data cleaning, feature encoding, and the implementation of Random Forest and XGBoost algorithms after data balancing with SMOTE. The findings reveal that the XGBoost algorithm achieves a prediction accuracy of 82%, outperforming Random Forest with 81%. Key factors influencing churn include Contract, TotalCharges, and tenure. The study concludes by emphasizing the significance of contract flexibility and the need to prioritize customers with high total costs or extended subscription periods to reduce churn rates. Future research is encouraged to investigate alternative methods for handling data imbalance and to explore advanced machine learning algorithms to further enhance prediction accuracy and the effectiveness of customer retention strategies
Lung X-ray Image Similarity Analysis Using RGB Pixel Comparison Method
Get PDFThe high death rate caused by pneumonia and Covid-19 is still quite high. Based on data released by WHO, 14% of deaths in children under 5 years old are caused by pneumonia. One of the processes carried out to help the diagnosis process is to look at lung images using X-Ray images. To obtain information about normal lung X-Ray images, Pneumonia and Covid-19, calculations are carried out using the color difference in each pixel of the X-ray image. The calculation process will provide output in the form of numbers in units of 0 to 100. This is done to facilitate the process of identifying the similarity of each X-Ray image being compared. The research stages are carried out with stages starting from adjusting the image size, then by breaking down the pixel values of the two images being compared and the process of calculating the difference in value from each pixel with the same coordinates. After calculating a combination of 30,000 combinations using 300 x-ray images, the results obtained in the form of the level of similarity between normal x-ray images and pneumonia x-ray images are the highest with a similarity percentage of 80.06%. The combination of normal images and pneumonia images is 10,000 combinations using 100 normal x-ray images and 100 pneumonia x-ray images. Normal x-ray images and covid x-ray images have a similarity of 79.18%. The combination of normal images and covid images is 10,000 combinations. The combination uses 100 normal x-ray images and 100 covid x-ray images. Pneumonia x-ray images and covid x-ray images have the lowest similarity level of 78.87%. The combination of pneumonia x-ray images and covid x-ray images is 10,000 combinations. The data used in the combination are 100 pneumonia images and 100 covid images. From the test results, the information obtained was that Accuracy was worth 0.54, Precision was worth 0.54, Recall was worth 0.59 and F1-score was worth 0.56.Tingginya angka kematian yang disebabkan oleh pneumonia dan covid-19 masih cukup tinggi. Berdasarkan data yang dirilis oleh WHO, 14% kematian pada anak dibawah 5 tahun, disebabkan oleh Pneumonia. Salah satu proses yang dilakukan untuk membantu proses diagnosis adalah dengan melihat citra paru-paru menggunakan citra X-Ray. Untuk mendapatkan informasi mengenai citra X-Ray paru-paru normal, Pneumonia dan Covid-19, dilakukan penghitungan menggunakan selisih warna pada masing-masing pixel citra X-ray. Proses penghitungan akan memberikan keluaran berupa angka dalam satuan 0 hingga 100. Hal ini dilakukan untuk memudahkan proses identifikasi kemiripan setiap citra X-Ray yang dibandingkan. Tahapan penelitian dilakukan dengan tahapan yang dimulai dari penyesuaian ukuran citra, kemuddian dengan memecah nilai pixel kedua citra yang dibandingkan dan proses penghitungan selisih nilai dari masing-masing pixel dengan koordinat yang sama. Setelah melakukan perhitungan kombinasi sebanyak 30.000 kombinasi dengan menggunakan 300 citra x-ray, maka didapatkan hasil berupa tingkat kemiripan antara citra x-ray normal dan citra x-ray pneumonia adalah yang tertinggi dengan persentase kemiripian sebesar 80.06%. Kombinasi antara citra normal dan citra pneumonia adalah 10.000 kombinasi dengan menggunakan 100 citra x-ray normal dan 100 citra x-ray pneumonia. Citra x-ray normal dan citra x-ray covid memiliki kemiripan sebesar 79.18%. Kombinasi antara citra normal dan citra covid adalah 10.000 kombinasi. Kombinasi menggunakan 100 citra x-ray normal dan 100 citra x-ray covid. Citra X-ray pneumonia dan citra X-ray covid memiliki tinggkat kemiripan terendah sebesar 78.87%. Kombinasi antara citra X-ray pneumonia dan citra X-ray covid adalah 10.000 kombinasi. Data yang digunakan dalam kombinasi yaitu 100 citra pneumonia dan 100 citra covi