Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Prediction of Cyberbullying in Social Media on Twitter Using Logistic Regression
Get PDFAs cases of cyberbullying on social media increase, there is a need for efficient measures to detect the vice. This research aims to establish the application of machine learning algorithms in analyzing text on social media to determine potentially harmful comments using logistic regression. The first and most important research question of this study is to assess the extent to which the model is capable of correctly identifying the comments that contain features of cyberbullying and those that do not. The data set included comments from different social media sites and was preprocessed before further analysis was conducted on it. Exploratory Data Analysis was applied in the study to establish relationships and textual features with bullying behavior. As with any other model, after training and testing the model, the results were analyzed using parameters like precision, precision, gain, and F1 statistics. The outcomes of this study revealed that the use of logistic regression models can give a fairly satisfactory level of accuracy in identifying cyberbullying. In light of this, this study underscores the need to use machine learning algorithms to minimize negative actions in cyberspace
Lung Segmentation in X-ray Images of Tuberculosis Patients Using U-Net with CLAHE Preprocessing
Get PDFTuberculosis (TB) is an infectious disease that commonly affects the lungs and remains one of the leading causes of death from infectious diseases. Early detection is essential to prevent further spread and organ damage. Chest X-ray images are one of the main methods for diagnosing TB, but image quality is often affected by low contrast and noise. This study proposes the application of Contrast Limited Adaptive Histogram Equalization (CLAHE) method to improve X-ray image quality, combined with U-Net deep learning architecture for lung segmentation in X-ray images of tuberculosis patients. U-Net was chosen due to its excellent capability in medical image segmentation, thanks to its architectural structure that has encoder-decoder with skip connections, which allows the model to retain detailed information on high-resolution images, even on complex and noisy data. Experimental results using the Shenzhen and Montgomery datasets show that the U-Net model with CLAHE achieves Pixel Accuracy 97.96%, Recall 94.93%, Specificity 98.97%, Dice Coefficient 95.87%, and Jaccard Index (IoU) 92.07%
Real-Time Chinese Chess Piece Character Recognition using Edge AI
Get PDFThis research focuses on developing a character analysis system on Chinese chess pieces (xiangqi) using computer vision technology with the deep learning framework PyTorch. The system is designed to detect and interpret text written on chess pieces in real time, making it easier for players to identify the function of each piece. The implementation is done using a web camera and can be applied to embedded devices such as Jetson Nano. This research aims to develop an automatic recognition system that can help players better understand the game of xiangqi by identifying characters on pieces in real time. The test results show that the system successfully recognized 14 pieces correctly. The system developed using Jetson Nano can directly process image data with a processing time of 0.0222 seconds. This data is obtained from the average of each FPS image from the web camera
Polynomial Integrated PLS Regression for Predicting Corrosion Inhibition Efficiency of Ionic Liquids
Get PDFCorrosion degrades and weakens metal surfaces, leading to structural failure and significant safety hazards across various sectors. Data driven machine learning offers a rapid, cost-effective alternative to the expensive and time consuming traditional experimental methods by predicting inhibitor performance computationally. This study addresses the challenge of accurately predicting corrosion inhibition efficiency (CIE) of ionic liquid compounds. Integration of a polynomial function, especially in higher degrees, inevitably grows the dimensionality and escalates multicollinearity, but it captures deeper nonlinear interactions that the original variables alone would miss. To counterbalance this curse of dimensionality, Partial Least Squares (PLS) Regression was applied after polynomial integration to project the high-dimensional variables into a smaller set of predictors. Besides PLS, Gradient Boosting Regressor (GBR) and Support Vector Regressor (SVR) models were also developed to establish baseline performance. Although these polynomial integrated models outperformed their baseline version, the Polynomial Integrated PLS outperformed their predictive performance, yielding R2, RMSE, and MAPE of 0.73, 4.730, and 3.73%, respectively. The result of this study highlights that the integration of a polynomial function can improve the predictive performance of PLS for corrosion inhibitors
Enhancing Eye Diseases Classification Using Imbalance Training & Machine Learning
Get PDFThis research aims to evaluate the effectiveness of various machine learning algorithms in classifying eye diseases based on retinal images. The dataset comprises four categories of eye diseases: Cataract, Diabetic Retinopathy, Glaucoma, and Normal. The feature extraction method employed a transfer learning approach using ResNet50, followed by SMOTE for data balancing, PCA for dimensionality reduction, and normalization for scaling data consistently. Eleven machine learning models were evaluated, including basic algorithms, ensemble methods, and neural networks. The evaluation utilized metrics such as accuracy, precision, recall, and F1-score. K-Fold Cross Validation is also employed to observe all models\u27 generalisation. The results revealed that the XGBoost algorithm achieved the highest performance with an accuracy of 92.03%, followed by LightGBM 91.88% and MLP 91.50%. K-Fold Validation also improved the MLP performance, which achieved an average accuracy of 91.94% with a standard deviation of 0.0178. This study successfully enhanced classification accuracy compared to previous studies and shows significant potential for clinical applications in resource-limited environments.This research aims to evaluate the effectiveness of various machine learning algorithms in classifying eye diseases based on retinal images. The dataset comprises four categories of eye diseases: Cataract, Diabetic Retinopathy, Glaucoma, and Normal. The feature extraction method employed a transfer learning approach using ResNet50, followed by SMOTE for data balancing, PCA for dimensionality reduction, and normalization for scaling data consistently. Eleven machine learning models were evaluated, including basic algorithms, ensemble methods, and neural networks. The evaluation utilized metrics such as accuracy, precision, recall, and F1-score. K-Fold Cross Validation is also employed to observe all models\u27 generalisation. The results revealed that the XGBoost algorithm achieved the highest performance with an accuracy of 92.03%, followed by LightGBM 91.88% and MLP 91.50%. K-Fold Validation also improved the MLP performance, which achieved an average accuracy of 91.94% with a standard deviation of 0.0178. This study successfully enhanced classification accuracy compared to previous studies and shows significant potential for clinical applications in resource-limited environments
Which Works Better for Kampus Merdeka MSIB: Ads or Content?
Get PDFThis study aims to evaluate the influence of social media advertising and content marketing on the brand awareness of the Kampus Merdeka MSIB program at Infinite Learning. The research uses a quantitative approach with non-probability sampling through purposive sampling techniques. A total of 160 students who had been exposed to Infinite Learning\u27s advertisements or content on Instagram participated as respondents, and data were collected through an online questionnaire. Data analysis was carried out using the SEM-PLS method. The results indicate that social media advertising has a positive and significant effect on brand awareness, while content marketing has a positive but not significant effect. Although the individual effects differ, the combination of both strategies shows a significant influence on increasing brand awareness. These findings suggest that the synergy between advertising and content strategies on social media is more effective than applying them separately in building brand awareness for Infinite Learning’s MSIB program
ECONOMIC VALUE ADDED, INVESTMENT OPPORTUNITY SET DAN KEBIJAKAN DIVIDEN
Get PDFThe pandemic has had a significant impact on the global capital market. The property and real estate sector is one of the sectors that has been dramatically affected by the pandemic. This is evident from the significant decline in stock trading volume in the property and real estate sector during the pandemic, compared to the previous year. This certainly attracts investors\u27 attention to dividend policies and investment opportunities, considering several key factors, including Economic Value Added (EVA) and Investment Opportunity Set (IOS). This study aims to examine the economic added value, determine investment opportunities, and analyze dividend policies in the property and real estate sector on the Indonesian stock exchange during the 2019-2020 period. The sample determination employed a purposive sampling method, resulting in a sample of 52 companies. The analysis technique used is multiple linear regression, utilizing SPSS 25 as the statistical software. Based on the research analysis, it was found that, partially, the EVA and IOS variables had an effect, albeit not significant, on dividends, both partially and simultaneously.Pandemi berdampak cukup signifikan terhadap pasar modal global. Sektor property dan real estate merupakan salah satu sektor yang sangat terdampak oleh pandemi. Hal ini terlihat pada volume perdagangan saham sektor property dan real estate dari masa pandemi yang menurun drastis dari tahun tahun sebelumnya. Keadaan ini tentunya menarik perhatian investor mengenai kebijakan dividen dan peluang investasi, seperti beberapa faktor diantaranya Economic Value Added (EVA) dan Investment Opportunity Set (IOS). Penelitian ini bertujuan untuk menguji economic value added, investment opportunity set dan kebijakan dividen pada sektor property dan real estate di bursa efek Indonesia pada periode 2019-2020. Penentuan sampel menggunakan metode purposive sampling dan diperoleh 52 perusahaan. Teknik analisis yang digunakan adalah analisis regresi linier berganda dengan menggunakan SPSS 25 sebagai alat pengujian. Berdasarkan hasil dari analisis penelitian ditemukan bahwa secara parsial variabel EVA dan IOS berpengaruh namun tidak signifikan terhadap kebijakan dividen secara parsial dan simultan
The Influence of Internet Service Provider (ISP) Service Quality in Indonesia on Consumer Loyalty Through Internet User Profiles
Get PDFThis study examines the effect of the quality of internet service providers (ISP) in Indonesia on consumer loyalty through the internet user profile. The population in this study are internet users and consumers (Telkomsel, Indosat, and XL) who have used the service provider for at least 1 year. The respondents used were 300 respondents located in Batam City, Indonesia. The sampling technique used was purposive method and quota sampling. This study aims to examine the dimensions of ISP service quality on consumer loyalty based on user groups (Light, Medium, and Heavy User). The data analysis technique used is SEM (Structural Equation Modeling) with SmartPLS, AMOS, SPSS, and JASP. The results of this study indicate that ISP service quality has a positive and significant effect on consumer loyalty in Indonesia. Meanwhile, the largest group of internet users is medium users.Penelitian ini menguji tentang Pengaruh Kualitas Layanan Internet Service Provider (ISP) di Indonesia Terhadap Loyalitas Konsumen Melalui Internet User Profile. Populasi dalam penelitian ini adalah para pengguna dan konsumen internet (Telkomsel, Indosat, dan XL) yang telah menggunakan jasa provider minimal selama 1 tahun. Adapun responden yang digunakan yaitu sejumlah 300 reponden yang berlokasi di Kota Batam, Indonesia. Teknik pengambilan sampel yang digunakan adalah dengan metode purposive dan quota sampling. Penelitian ini bertujuan untuk menguji dimensi kualitas layanan ISP terhadap loyalitas konsumen berdasarkan pada kelompok penggunanya (Light, Medium, dan Heavy User). Teknik analisis data yang digunakan yaitu SEM (Structural Equation Modeling) dengan SmartPLS, AMOS, SPSS, JASP. Hasil dari penelitian ini menunjukkan bahwa kualitas layanan ISP berpengaruh positif dan beberapa signifikan terhadap loyalitas konsumen di Indonesia. Adapun kelompok pengguna internet yang paling besar adalah medium user
Analisis Risiko Pada Proses Rigging dan Lifting Menggunakan Metode HIRARC di Production Platform Module Erection Area PT. McDermott Indonesia
No full textPT. McDermott Indonesia is a company engaged in offshore construction services. Workplace accidents in the offshore construction industry are often due to poor weather conditions, inadequate work locations, equipment failures, and incompetent employees. This study aims to determine the risk level in the rigging and lifting process using the HIRARC method as an effort to achieve zero accidents in the Production Platform Module Erection Area. This research uses a descriptive qualitative approach. The research subjects consist of 5 informants: a rigging superintendent, an HSE specialist, a rigger, a crane operator, and a mechanic foreman, assisted by interview guidelines and a risk matrix. The study results indicate 21 potential hazards across 4 work stages and 10 work processes, categorized into 6 low risk, 8 medium risk, 13 high risk, and 4 extreme risk. Risk control is carried out by HSE procedures and the lifting plan, and it is recommended to use PPE consistently. In conclusion, the risk analysis in the rigging and lifting process has been conducted according to procedures, but updates to the TRA are necessary to reduce increasing risks. It is recommended to enhance employee awareness and discipline regarding the importance of implementing HSE measures.PT. McDermott Indonesia merupakan perusahaan yang bergerak di bidang jasa konstruksi lepas pantai. Kecelakaan kerja pada industri konstruksi lepas pantai sering kali disebabkan oleh kondisi cuaca yang buruk, lokasi kerja yang kurang memadai, kegagalan peralatan, dan karyawan yang tidak kompeten. Penelitian ini bertujuan untuk mengetahui tingkat risiko pada proses rigging dan lifting dengan menggunakan metode HIRARC sebagai upaya untuk mencapai zero accident di Production Platform Module Erection Area. Penelitian ini menggunakan pendekatan kualitatif deskriptif. Subjek penelitian terdiri dari 5 orang informan yaitu seorang pengawas rigging, seorang HSE specialist, seorang rigger, seorang operator crane, dan seorang mandor mekanik, yang dibantu dengan pedoman wawancara dan matriks risiko. Hasil penelitian menunjukkan 21 potensi bahaya pada 4 tahapan kerja dan 10 proses kerja, yang dikategorikan menjadi 6 risiko rendah, 8 risiko sedang, 13 risiko tinggi, dan 4 risiko ekstrim. Pengendalian risiko dilakukan dengan prosedur K3LH dan lifting plan, serta disarankan untuk menggunakan APD secara konsisten. Kesimpulannya, analisis risiko pada proses rigging dan lifting telah dilakukan sesuai prosedur, namun perlu dilakukan pembaharuan pada TRA untuk mengurangi risiko yang meningkat. Disarankan untuk meningkatkan kesadaran dan kedisiplinan karyawan mengenai pentingnya penerapan tindakan-tindakan K3
Financial Ratio Analysis and Health Level of PT Wijaya Karya (Persero) Tbk Based on Altman Z-Score
Get PDFThis study employs a quantitative descriptive approach to evaluate the financial performance and health of PT Wijaya Karya (Persero) Tbk. It involves analyzing the company\u27s financial ratios, including profitability, liquidity, activity, and solvency ratios, using the financial statements available on the IDX website from 2020-2023. The company\u27s health is assessed using the four-variable version of the Altman Z-Score method. The results indicate a declining trend in liquidity ratios, suggesting a reduced ability to meet short-term obligations. Fluctuations in solvency ratios imply increased risk or development through higher borrowing. Activity ratios reflect an overall increase, but with fluctuations in fixed asset utilization and collection efficiency. Profitability ratios demonstrate a downward trend over time. According to the Altman Z-Score analysis, the company faces a high risk of bankruptcy due to its vulnerable health level, necessitating immediate action to address this critical issue.Penelitian ini menggunakan pendekatan deskriptif kuantitatif untuk mengevaluasi kinerja keuangan dan kesehatan PT Wijaya Karya (Persero) Tbk. Penelitian ini dilakukan dengan menganalisis rasio keuangan perusahaan, termasuk rasio profitabilitas, likuiditas, aktivitas, dan solvabilitas, dengan menggunakan laporan keuangan yang tersedia di situs web BEI dari tahun 2020-2023. Kesehatan perusahaan dinilai dengan menggunakan metode Altman Z-Score versi empat variabel. Hasilnya menunjukkan tren penurunan rasio likuiditas, yang menunjukkan berkurangnya kemampuan untuk memenuhi kewajiban jangka pendek. Fluktuasi dalam rasio solvabilitas menyiratkan peningkatan risiko atau perkembangan melalui pinjaman yang lebih tinggi. Rasio aktivitas mencerminkan peningkatan secara keseluruhan, tetapi dengan fluktuasi dalam pemanfaatan aset tetap dan efisiensi penagihan. Rasio profitabilitas menunjukkan tren penurunan dari waktu ke waktu. Menurut analisis Altman Z-Score, perusahaan menghadapi risiko kebangkrutan yang tinggi karena tingkat kesehatannya yang rentan, sehingga memerlukan tindakan segera untuk mengatasi masalah kritis ini