Sinkron : jurnal dan penelitian teknik informatika
Not a member yet
1261 research outputs found
Sort by
Comparison of K-Nearest Neighbor, Naive Bayes, Random Forest Algorithms for Obesity Prediction
No full textObesity is a global health problem that continues to increase and has serious impacts on physical and mental health. This research aims to predict a person's obesity status based on certain attributes using the K-Nearest Neighbor (KNN), Naive Bayes, and Random Forest algorithms. The dataset used was taken from the Kaggle platform with 2,111 data and 16 attributes, including gender, age, weight, height, frequency of consumption of high-calorie foods, physical activity, and water and vegetable consumption patterns. The research process follows the data mining stages, including business understanding, data understanding, data preparation, modeling, evaluation, and documentation. Experiments were carried out using RapidMiner with a cross-validation technique using 10 folds to measure overall model performance. The research results show that the Random Forest algorithm performs best in predicting obesity status compared to K-NN and Naive Bayes. Model evaluation using accuracy, precision, recall, and F1-score metrics shows significant results in distinguishing obesity categories. It is hoped that this research can contribute to the development of a machine learning-based health prediction system that can be used to support decision-making in the prevention and management of obesity
The Enterprise Architecture for Enhanced Mutual Fund Service Integration in Digital Channel in the Banking Industry
No full textThe investment awareness among the Indonesian populace is rising, particularly during the Covid-19 pandemic. Mutual funds are the most prevalent investment vehicle due to their accessibility, requiring minimal cash, and offering competitive returns, as they are managed by seasoned investment professionals. This is due to the relative ease of investing in this instrument with accessible capital and competitive returns, as fund management is conducted by seasoned investment managers. Conversely, the rapid advancement and proliferation of technology present new challenges for firms in the financial sector, particularly in banking and financial technology, as they strive to innovate and provide convenient services that cater to the needs of customers and investors. This research seeks to develop Enterprise Architecture for the integration of Mutual Fund services into digital channels via mobile banking, utilizing the TOGAF framework as the foundational design approach, supplemented by SWOT analysis to assess the strategic position and market potential. The research employed a qualitative methodology utilizing the ArchiMate program to illustrate diagrams in Enterprise Architecture (EA) across the public banking sector. The research indicated that architectural design can facilitate improved data access and enhance adaptability to technology and market advancements, hence removing inefficiencies and streamlining the review process. In conclusion, the application of EA in the incorporation of mutual fund services into the digital banking platform will optimize the company's performance processes to attain objectives, while also enhancing agility in evaluating technological risks to facilitate the monitoring, metrics, and analysis of information technology, thereby ensuring the achievement of business goals
Automated Attendance System for Contract-Based Employees at Purwakarta Communication and Informatics Agency
No full textAttendance is a crucial aspect of administrative management in both companies and institutions. However, manual attendance systems have several drawbacks, including disorganized data, minimal data control leading to a high potential for data fraud, and vulnerability to data loss or damage. These limitations underscore the need for more efficient and reliable attendance management solutions. This research addresses the challenges of manual attendance tracking for non-ASN (Aparatur Sipil Negara) staff at the Communication and Informatics Agency (Dinas Komunikasi dan Informatika (DISKOMINFO)) of Purwakarta Regency by developing a user-friendly web-based attendance system. The system leverages the PHP programming language and MySQL database to efficiently record and manage attendance data. The Waterfall method is used as the development framework, ensuring a structured and systematic approach. The system incorporates features designed with Unified Modelling Language (UML) to simplify attendance recording for staff and administrators, including online check-in/check-out, real-time attendance tracking, and automated report generation. Evaluation results demonstrate that the system significantly improves attendance accuracy, reduces administrative burden, and enhances overall efficiency within the office. This research highlights the importance of embracing technology to modernize administrative processes and improve operational effectiveness in government organizations. The implementation of this technology was also tested for its effectiveness using black box testing and the usability scale system
Beyond Traditional QoS Management- Harnessing Machine Learning for Predictive Network Service Optimization
No full textQuality of Service (QoS) is a fundamental aspect of modern computer networks, directly influencing performance and user experience. Key parameters such as latency, throughput, packet loss, and jitter play crucial roles in determining network efficiency. Traditional QoS management approaches, often rule-based or heuristic-driven, lack adaptability to dynamic network conditions. This study explores the application of machine learning techniques to predict QoS using historical network data, enabling proactive network optimization. We employ multiple predictive models, including linear regression, random forest, and deep learning, to analyze network performance trends and forecast QoS degradation. Experimental results demonstrate that machine learning significantly enhances prediction accuracy compared to conventional methods, allowing for more effective resource allocation and congestion control. The findings highlight the potential of data-driven approaches in real-time network management, reducing latency fluctuations and improving service reliability. Moreover, deep learning models outperform traditional statistical techniques in recognizing complex patterns within network data, making them a promising solution for next-generation network optimization. The proposed methodology not only improves predictive accuracy but also offers a scalable framework for automated QoS management in cloud computing, IoT, and 5G environments. Future work will focus on refining model generalization across diverse network conditions and integrating federated learning for privacy-preserving QoS predictions. This research underscores the transformative role of machine learning in enhancing network service quality and operational efficiency
The Influence of Service Quality Dimensions and E-Commerce Image on User Satisfaction Levels
No full textThis study investigates how service quality and brand image influence user satisfaction on Shopee, Indonesia’s leading e-commerce platform, amid persistent operational challenges like delivery delays and inefficient return processes. Using a quantitative approach, data was collected via validated questionnaires and analysed through multiple regression, with diagnostic tests ensuring model reliability. Results show both service quality and brand image significantly boost satisfaction, with brand image having nearly twice the impact. The model’s high explanatory power, with adjusted R² is 0.943, confirms their dominant role in shaping user experiences. These findings align with prior retail and hospitality studies but challenge traditional service-dominant logic by highlighting brand perception as the stronger driver in digital commerce. For practitioners, this underscores the need to prioritize brand-building strategies, such as trust signals and influencer partnerships, while maintaining service standards. However, the study’s exclusive focus on Shopee limits generalizability, calling for future cross-platform comparisons. Additional research should employ longitudinal designs, examine cultural moderators, and apply advanced analytics to uncover deeper satisfaction dynamics. By validating the interplay of service quality and brand equity in e-commerce, this study advances theoretical discourse while offering actionable insights for platforms seeking to enhance customer satisfaction in competitive digital marketplaces
Hyperparameter-Tuned Artificial Neural Networks for Early Stunting Prediction in Toddlers
No full textThe growing accessibility of varied health data requires the creation of efficient and practical techniques for deriving actionable insights, particularly for the early identification of severe health issues. This study tackles the issue of recognizing stunting—a disorder with enduring health consequences—among children under five by employing Artificial Neural Networks (ANN) with hyperparameter optimization by GridSearchCV. The dataset, sourced from Kaggle, comprises 121,000 records detailing age, gender, height, and nutritional status according to WHO standards. Critical hyperparameters, including dropout rate, batch size, and epochs, were optimized using a five-fold cross-validation approach within GridSearchCV, ensuring a robust model that reduces overfitting and generalizes well to new data. The findings demonstrate a notable performance improvement, as the optimized ANN model attained an accuracy of 99%, exceeding the baseline model's 98%. Enhancements in accuracy, recall, and F1-score across the four stunting classifications—normal, stunted, severely stunted, and tall—underscore the improved specificity and sensitivity of the optimized model. This research demonstrates the efficacy of hyperparameter tuning in ANN for stunting prediction, offering a reliable tool for early intervention in malnutrition management
Comparative Analysis of LSTM, GRU, and Bi-LSTM Deep Learning Models for Time Series Cryptocurrency Price Forecasting
No full textCryptocurrency is a highly volatile digital asset that requires accurate predictive methods. This study compares the performance of three deep learning architectures LSTM, GRU, and Bi-LSTM in forecasting the prices of Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB) using univariate historical data. Evaluation was conducted through regression metrics (RMSE and MAPE) and classification of price movement into five categories, ranging from very bearish to very bullish, assessed using a confusion matrix. The results show that GRU performed best for BTC (RMSE 974.72, MAPE 1.18%), while Bi-LSTM outperformed others for ETH and BNB (RMSE 43.19 and 6.83; MAPE 1.16% and 1.08%) and achieved the highest classification accuracy (55% and 52%). However, overall classification accuracy remains low, reflecting the complexity of cryptocurrency price patterns. The study is limited by its univariate approach without incorporating external variables. Its contribution lies in combining regression and classification evaluation, and it recommends exploring multivariate and ensemble models in future research
Bibliometric Mapping and Trend Analysis of Beta Regression Modeling: A Decade of Development (2015–2024)
No full textBeta regression is a statistical model designed to handle dependent variables that assume values within the open interval (0, 1), such as rates, proportions, or percentages. The study aimed to determine the development of beta regression over the last 10 years with a bibliometric approach. The source of the article database used comes from the Scopus website. The tool used for analysis is R software with a bibliometrix package. The results of this study show that there are 293 articles published in the Scopus Journal. Research develops in various research fields. The author with the most articles is Cribari-Neto, F., with the most significant number of documents, i.e., 12. According to the author's country of origin related to the beta regression method, Brazil has the most countries, while Indonesia is in 12th place. Therefore, research on beta regression still has excellent potential to continue to be developed
Integrating SMOTE with XGBoost for Robust Classification on Imbalanced Datasets: A Dual-Domain Evaluation
No full textClass imbalance is one of the main challenges in classification problems, as it can reduce the model's ability to accurately identify minority classes and negatively impact the overall reliability of predictions. In response to this problem, this study proposes an integrated approach combining SMOTE and XGBoost to improve classification performance on imbalanced data. This approach aims to evaluate the impact of oversampling techniques on prediction accuracy and model sensitivity to class distribution. The evaluation was conducted using two public datasets representing different domains and different amounts of data, namely Spambase and Diabetes, to assess the effectiveness and generalization of the applied approach. The experimental results show that this integrated model consistently outperforms traditional comparison algorithms, with an F1 score of 0.94 and ROC-AUC of 0.98 on the Spambase dataset and ROC-AUC of 0.83 on the Diabetes dataset, with a good balance between precision and recall. The 10-fold cross-validation technique was applied to ensure objective performance estimates free from random data splitting bias. Additionally, this study highlights the importance of selecting appropriate evaluation metrics in the context of imbalanced data, as single accuracy often provides a misleading performance picture. This study makes a significant contribution by providing a benchmark for comparing the effectiveness of SMOTE-XGBoost integration using two different datasets, accompanied by rigorous cross-validation. These findings reinforce the position of integrating data preprocessing strategies and ensemble learning as a competitive and adaptive solution for addressing class imbalance challenges in data-driven classification systems
Indonesian Public Sentiment Toward Electric Vehicles: Analysis of Social Media Data
No full textThe development of electric vehicles (EVs) in Indonesia has progressed significantly, supported by government subsidies for Battery-Based Electric Motor Vehicles. These subsidies have sparked mixed public reactions that some support them due to environmental benefits and pollution reduction, while others oppose them for various reasons. Social media platform X serves as a valuable source for gauging public opinion, though analyzing such data manually can be complex. To address this, sentiment analysis, particularly using the Support Vector Machine (SVM) method, offers an efficient solution. This study analyzes 23,031 Indonesian-language tweets from social media platform X, collected between October 2023 and July 2024, using SVM for sentiment classification. The best-performing model, with parameter C = 0.5 and without stemming, achieved an accuracy of 84.98%. The findings suggest that Indonesians generally view electric vehicles positively, with more favorable sentiments than negative ones. This study offers implications across methodological, industrial, and policy domains. Word cloud analysis further supports this, highlighting public support in areas such as pricing, infrastructure, and environmental impact. However, the study also identifies key concerns, including issues around subsidies, taxes, vehicle durability, battery types, and import regulations. Overall, the research provides meaningful insights into the diverse perspectives of Indonesian citizens regarding EVs, helping to inform future policy and development strategies