Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Enhancing Interpretable Multiclass Lung Cancer Severity Classification using TabNet
No full textLung cancer poses a significant global mortality challenge, with early clinical detection hindered by non-specific symptoms making accurate diagnosis dependent on extracting subtle patterns from often complex medical tabular data. Traditional machine learning approaches often fall short in capturing intricate patterns within such heterogeneous datasets, hindering effective clinical decision support. This research introduces TabNet, an interpretable deep learning architecture, for multiclass lung cancer severity prediction (low, medium, high). Utilizing the Kaggle Lung Cancer dataset, our methodology leverages TabNet\u27s unique attention-based feature selection for end-to-end processing of tabular data, enabling adaptive identification of key predictors and crucial model interpretability. To effectively assess its predictive capabilities and ensure robust performance, the model was trained with default configurations and validated through stratified 5-fold cross-validation, achieving outstanding performance on the test set: 98.50% accuracy, a 0.98 F1-score, and a 0.9996 macro-AUC-ROC. Beyond its robustness, confirmed by stable learning curves, interpretability analysis highlighted \u27Genetic Risk\u27 and \u27Shortness of Breath\u27 as dominant factors. Our results underscore TabNet\u27s efficacy as a reliable, robust, and inherently interpretable solution, offering significant potential to improve the precision and transparency of lung cancer severity assessment in clinical practice
Optimizing F1 Tyre Performance Prediction with SVC, XGBoost, and Optuna For Dutch GP 2022
No full textFormula 1 has evolved into a data-centric sport where strategic decisions, particularly tire compound selection (Soft, Medium, Hard), are critical for success. The ability to accurately identify a competitor\u27s compound from observable telemetry data offers a significant strategic advantage, yet the predictive signals are subtle and difficult to distinguish. This study implements and compares two distinct machine learning methodologies to classify F1 tyre compounds using telemetry data from the 2022 Dutch Grand Prix. First, a baseline model was established using standard dynamic features (e.g., avg_speed, avg_rpm). While this approach confirmed the superiority of XGBoost over SVC, it yielded a modest accuracy of 67.99% and revealed a critical deficiency: a failure to reliably identify the HARD compound, registering a poor F1-score of 0.57. To address these limitations, an advanced methodology was developed, integrating hybrid feature engineering (e.g., LapTime, SectorTime, TyreLife) with deep hyperparameter optimization via Optuna. This enhanced approach resulted in a significantly more robust XGBoost model, achieving a final, stable accuracy of 77.34%. More importantly, it solved the baseline\u27s primary flaw, increasing the F1-score for the critical HARD compound by 36.8% to 0.78. A feature importance analysis confirmed this methodological shift, as the most dominant predictors changed from the baseline\u27s generalized avg_speed to the advanced model\u27s outcome-based features (LapTime, Sector3Time). The findings definitively conclude that while algorithm selection is important, the most critical factor for this task is the quality of feature engineering. Integrating outcome-based and strategic-context features is essential to successfully extracting the subtle performance signatures that differentiate F1 tyre compounds
Development Of A Collaborative Recommendation System Based on Singular Value Decomposition (SVD) on E-Commerce Data
No full textRecommendation systems (RS) are vital tools for mitigating information overload and data sparsity challenges in modern e-commerce platforms. This study focuses on developing and evaluating a Collaborative Filtering (CF) model utilizing Singular Value Decomposition (SVD) as a Matrix Factorization technique, applied to the publicly available E-commerce dataset. The dataset, comprising nine interconnected transactional tables, presents significant data sparsity due to limited explicit user ratings relative to the vast product catalog. The SVD model was implemented to decompose the highly sparse User-Item interaction matrix into lower-rank latent factor matrices, thereby capturing underlying purchasing patterns and user preferences. The model\u27s performance was rigorously validated using k-fold cross-validation and assessed via standard accuracy metrics: Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The experimental results demonstrated high predictive accuracy, achieving an RMSE of 1.25 and an MAE of 0.98. These findings indicate that the SVD model effectively overcomes the sparsity challenge inherent in large-scale e-commerce transactional data, providing robust prediction capabilities that surpass established industry benchmarks (e.g., RMSE » 1.31, MAE » 1.04 found in similar studies). The successful implementation validates SVD as a highly effective approach for generating personalized, high-quality product recommendations, offering substantial business implications for enhancing customer engagement and maximizing Average Order Value (AOV
Classification of Tumor and Normal Tissue Gene Expression in Lung Adenocarcinoma Using Support Vector Machine and Gaussian Process Classification
No full textLung adenocarcinoma (LUAD) is a major cause of cancer-related mortality worldwide. This study aims to identify potential LUAD biomarkers and develop robust classification models using the GSE151101 microarray dataset. Preprocessing included RMA normalization, ComBat batch-effect correction, and feature filtering based on annotation completeness, variability, and statistical significance. Support Vector Machine (SVM) and Gaussian Process Classification (GPC) models were constructed, with the polynomial GPC model achieving the best performance (accuracy 97.92%; F1-score 97.96%). Repeated 10-fold cross-validation confirmed its stability (mean accuracy 96.88%, SD ±1.97%, CV 2.03%), outperforming linear SVM, GPC-RBF, and Multiple Kernel Learning (MKL). Functional enrichment analysis showed that key discriminative genes; CDH13, CDKN2A, BCL2L11, MYL9, COL1A1, and AKT3; were enriched in pathways related to epithelial–mesenchymal transition, extracellular matrix remodelling, focal adhesion, PI3K/AKT signalling, and cell-cycle regulation, all of which are central to LUAD progression. In general, polynomial-kernel GPC is a stable and useful way to classify transcriptomes and rank biomarkers. Nevertheless, the translational potential of these signatures requires further validation in independent and clinically controlled cohorts
The Effect of Price and Product Quality on Customer Loyalty with Customer Satisfaction as a Mediating Variable: Case Study at Luargaris Coffee & Kitchen
No full textThis research investigates how price and product quality influence customer loyalty, with customer satisfaction acting as a mediating variable, focusing on Luargaris Coffee & Kitchen. A quantitative approach was employed, using purposive sampling, and the study involved 372 respondents. This research primarily relies on primary data, which was gathered directly from participants who are consumers of Luargaris Coffee & Kitchen. The primary data was obtained through a carefully structured questionnaire designed specifically to address the research objectives. The dependent variable in this study is consumer loyalty, while the independent variables are price and product quality. Additionally, consumer satisfaction serves as the intervening variable. The data were analyzed through multiple linear regression, applying the SEM-PLS technique. The results indicate that product quality positively and significantly affects customer loyalty, mediated by customer satisfaction. In contrast, price and service quality did not show a significant impact on customers’ purchasing decisions. The findings highlight the importance of prioritizing product quality improvement as a fundamental strategy to enhance long-term customer satisfaction and foster customer loyalty
DETERMINASI TAX AVOIDANCE PADA INDUSTRI PERTAMBANGAN: BUKTI EMPIRIS DARI INDONESIA
No full textThis study aims to examine the influence of profitability, the corporate income tax rate, firm size, and independent commissioners on tax avoidance. The population in this research consists of mining sector companies listed on the Indonesia Stock Exchange (IDX) during 2018–2021. The sampling technique used is purposive sampling, resulting in a total of 82 research samples. This study employs secondary data derived from the companies\u27 financial statements published on the official IDX website. The panel data analysis results demonstrate that profitability has a positive and significant effect on tax avoidance. Meanwhile, the corporate income tax rate has a negative and significant effect on tax avoidance. Conversely, firm size and the proportion of independent commissioners do not have a significant effect on tax avoidance.Penelitian ini dilakukan dengan tujuan untuk menguji adanya pengaruh profitabilitas, tarif PPh Badan, ukuran perusahaan, dan komisaris independen terhadap tax avoidance. Populasi yang digunakan dalam penelitian ini yaitu perusahaan pertambangan yang terdaftar di bursa efek indonesia pada periode tahun 2018-2021. Teknik pengambilan sampel menggunakan metode purposive sampling sehingga didapatkan jumlah 82 sampel penelitian. Penelitian ini menggunakan data sekunder yaitu laporan keuangan perusahaan yang dipublikasikan pada website bursa efek Indonesia. Hasil pengujian dengan menggunakan data panel membuktikan bahwa variabel profitabilitas berpengaruh positif dan signifikan terhadap tax avoidance. Variabel tarif PPh Badan menunjukkan pengaruh negative dan signifikan terhadap tax avoidance. Variabel ukuran perusahaan dan komisaris independen tidak berpengaruh terhadap tax avoidance
Effectiveness Analysis of Fintech on Financial Performance and Banking Sustainability
No full textThe development of Financial Technology (Fintech) in Indonesia is quite high, especially in banking. Ease of access and service, speed of transaction processing, simple requirements that can be fulfilled at any time and from any location, and the ability to reduce operational costs are key considerations for banks to adopt it. But behind these advantages, the usage of Fintech causes concern because it is easily hacked and can be misused for fraudulent activities. This research aims to explore and understand the effectiveness of Fintech on Financial Performance and Banking Sustainability using the Unified Theory of Acceptance and Use of Technology (UTAUT). The research utilized primary data collected from eight informants, comprising one policymaker, one implementer, and six users. Data collection techniques are observation and in-depth interviews. Data analysis includes data reduction, data presentation, and conclusion. The research results stated that the service implementation of Fintech in improving financial performance is considered effective in terms of performance expectations, business expectations, and social influence, but not yet effective in terms of facilitating conditions. The service implementation of Fintech to support service sustainability is considered effective in terms of performance expectations, social influence, and facilitating conditions, but not yet effective in terms of business expectations
Implementasi Efek Warp Stabilizer dalam Pembuatan Video Profil Program Studi Teknik Mesin
Get PDFThe instability of video recordings is a common issue that often occurs during recording, especially when using limited equipment or in uncontrolled situations. This problem can become serious when it is not possible to re-record. Therefore, a solution is needed at the video editing stage. One existing solution is using the warp stabilizer effect in Adobe Premiere Pro. Based on research conducted on the profile video of the Mechanical Engineering Study Program, there were 8 recordings that experienced this issue according to the profile video creators. Alpha testing resulted in 6 video recordings being categorized as unstable. Then, beta testing yielded 95.33% in the unstable category, 89.37% in the very unstable category, and 91.36% overall, which means that the implementation of the warp stabilizer is very effective in reducing instability, making the profile video of the Mechanical Engineering Study Program more professional.Ketidakstabilan rekaman video adalah salah satu masalah yang sering terjadi saat melakukan perekaman terutama jika dilakukan dengan peralatan yang terbatas atau dalam situasi yang tidak terkendali. Masalah ini dapat menjadi serius ketika tidak memungkinkan untuk melakukan perekaman ulang. Oleh karena itu, diperlukan solusi pada tahap pengeditan video. Salah satu solusi yang ada yaitu menggunakan efek warp stabilizer di Adobe Premiere Pro. Berdasarkan penelitian yang dilakukan pada video profil Program Studi Teknik Mesin yang mana terdapat 8 rekaman yang mengalami masalah ini menurut para pembuatan video profil tersebut. Pengujian alpha menghasilkan 6 rekaman video yang masuk kategori tidak stabil. Kemudian pengujian beta menghasilkan 95,33% pada kategori tidak stabil, 89,37% pada kategori sangat tidak stabil, dan 91,36% secara keseluruhan yang artinya adalah implementasi warp stabilizer sangat efektif dalam mengurangi ketidakstabilan sehingga membuat video profil Program Studi Teknik Mesin menjadi lebih profesional
Development of Virtual Lab on Collision Dynamics Learning Object with Collision Algorithm Integration
Get PDFThe objective of this study is to evaluate the efficacy of a Virtual Lab employing a collision algorithm in enhancing students\u27 conceptual comprehension of collision dynamics, in comparison to traditional pedagogical approaches, within the context of physics education.The methodology employed in this study is as follows: The study employed an experimental approach, comprising a comparison between two groups: an experimental class that used the Virtual Lab, and a control class that utilised traditional teaching methods. Both groups were subjected to pre-tests to ascertain their existing level of understanding, after which post-tests were conducted to evaluate their knowledge after the instruction period. An independent t-test was employed to analyse the differences in post-test outcomes between the two groups.The results are as follows: The findings indicated a significant improvement in the experimental class\u27s understanding, with an average increase from the pre-test to the post-test of 33.89%, in comparison to a 30.74% improvement in the control class. The results of the t-test demonstrated a statistically significant difference (t = 4.32, p < 0.05), indicating that the Virtual Lab was more effective in enhancing conceptual comprehension. In conclusion, the Virtual Lab, based on the collision algorithm, has been demonstrated to be an effective tool for teaching collision dynamics, offering a more interactive and engaging experience than traditional methods. This study highlights the potential of technology-based learning tools to enhance physics education and recommends further development of Virtual Labs with interactive features to increase accessibility and understanding in diverse educational environments
Optimization of Distribution Routes Using the Genetic Algorithm in the Traveling Salesman Problem
Get PDFTransportation plays a vital role in business operations, as it is essential for product distribution to maintain profitability. Optimizing distribution routes is crucial to reducing transportation costs, travel time, energy usage, and resource allocation while maximizing efficiency. Micro-entrepreneurs, particularly settled retailers, often face challenges in determining optimal travel routes, resulting in inefficiencies in product distribution. This issue is classified as a Traveling Salesman Problem (TSP), which involves finding the shortest possible route connecting several locations before returning to the starting point. To address this problem, this study applies a two-step approach: the greedy algorithm to provide an initial solution and the genetic algorithm for further optimization. The research employs both manual calculations and MATLAB 2018A software to solve the TSP. Results demonstrate that the optimized route reduces the travel distance by 1,260 meters compared to the initial solution, highlighting significant improvements in operational efficiency.Transportation plays a vital role in business operations, as it is essential for product distribution to maintain profitability. Optimizing distribution routes is crucial to reducing transportation costs, travel time, energy usage, and resource allocation while maximizing efficiency. Micro-entrepreneurs, particularly settled retailers, often face challenges in determining optimal travel routes, resulting in inefficiencies in product distribution. This issue is classified as a Traveling Salesman Problem (TSP), which involves finding the shortest possible route connecting several locations before returning to the starting point. To address this problem, this study applies a two-step approach: the greedy algorithm to provide an initial solution and the genetic algorithm for further optimization. The research employs both manual calculations and MATLAB 2018A software to solve the TSP. Results demonstrate that the optimized route reduces the travel distance by 1,260 meters compared to the initial solution, highlighting significant improvements in operational efficiency