JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
Comparative Analysis of Robust Imputation Techniques for Enhancing Cervical Cancer Prediction with Missing Data
Get PDFHandling missing data is a critical challenge in machine learning applications, as it can significantly affect the accuracy and reliability of predictive models. Addressing this issue is crucial for developing robust systems that can deliver high-performance results. This study provides a comparative analysis of the robust imputation technique for cervical cancer prediction with incomplete information. This study has investigated the importance of robust imputation techniques, particularly Soft Imputer, in addressing missing data challenges and enhancing model performance. This study investigates the impact of various imputations across five distinct approaches: KNN imputer, PCA imputer, MICE imputer, XGBoost imputer, LightGBM imputer, and feature selection methods. These imputation data are tested on several machine learning models such as Random Forest (RF), Extreme Gradient Boosting (XGB), Decision Tree (DT), Support Vector Classifier (SVC), Logistic Regression (LR), Extra Trees Classifier (ETC), CatBoost Classifier, Stochastic Gradient Descent (SGD), and Gradient Boosting (GB) for improving classification accuracy of cervical cancer prediction. The evaluation reveals that the soft imputer method achieves a balanced and effective handling of missing data, significantly improving the reliability of the models. Among the tested methods, LightGBM and XGBoost deliver strong results, each achieving an average accuracy of 96.91%. MICE demonstrated the lowest average accuracy at 95.94%, although it still performs reliably in managing missing data. The findings provide valuable insights for enhancing predictive accuracy in future work by integrating advanced imputation strategies for high-dimensional and complex datasets
Detection of Oil Palm Fruit Ripeness through Image Feature Optimization using Convolutional Neural Network Algorithm
Get PDFThe increase in the need for raw materials for palm oil products in the form of food and non-food is felt by the people of Indonesia and other countries. For this reason, triggering oil palm farmers in Indonesia must be able to maximize their production. Currently, oil palm farmers in Indonesia still need help knowing the level of sustainability of oil palm fruit to maintain their production. This research was conducted to identify the maturity level of oil palm fruit using practical images for oil palm farmers in Indonesia. The Convolutional Neutral Network (CNN) algorithm is the research method used to identify pictures of oil palm fruit. The dataset collection comprised 400 images of oil palm fruits divided into three types of classes, namely images of raw, ripe, and rotten oil palm fruits. The dataset was taken from various internet sources, and photos were taken directly using a mobile phone camera according to a predetermined class. This study found that identifying the maturity level of oil palm fruit using the Convolutional Neural Network (CNN) algorithm obtained a high accuracy of 98% in the training process and 76% in the model testing process. The findings of this study can also inspire further research in optimizing image features and using the Convolutional Neural Network (CNN) algorithm more efficiently. This could include a reduction in model training time, the number of parameters, or the development of other techniques that improve algorithm performance
Canva-based Animation Comic Video Media in Informatics Learning at SMP Negeri 14 Padang
Get PDFThis research addresses challenges in conventional learning media, limited facilities, and suboptimal use of technology, which hinder student material mastery, motivation, and independence. The study aims to develop and validate Canva-based animated comic video learning media for Informatics subjects in class VIII at SMP N 14 Padang. Using the Research and Development (R&D) method with a 4D development model—define, design, develop, disseminate—primary data were collected from validators, teachers, and students. Descriptive and inferential analyses were employed to evaluate the validity and practicality of the media. The learning media achieved high validity scores: 0.963 for media design and 0.975 for material content. Expert evaluations highlighted the media’s effective visual design, systematic content presentation, and alignment with curriculum objectives. Practicality was confirmed with average scores of 97.04% from teachers and 93.14% from students, who appreciated its ease of use, accessibility across devices, and engaging, interactive features that support both independent and collaborative learning. This study underscores the importance of integrating technology into learning media to enhance education quality. Canva-based animated comic videos are not only applicable to Informatics but also have potential for adaptation to other subjects. The combination of visual, audio, and interactive elements fosters engaging, flexible, and impactful learning experiences for students. Future research could explore AI integration for personalized learning and broader testing across diverse student groups and subjects. This research provides a foundation for developing innovative, accessible, and inclusive technology-based learning tools to improve education quality in the digital era
Examining the Impact Factors Influencing Higher Education Institution (HEI) Students’ Security Behaviours in Cyberspace Environment
Get PDFThe Internet’s increasing connectivity through devices and systems, particularly with the Internet of Things (IoT), has expanded the threat landscape, making cybersecurity a constantly evolving field. Phishing is a common and emerging cyber-attack that attempts to deceive individuals and persuade them to disclose sensitive information, such as passwords, financial information, or personal data. Researchers have studied phishing extensively in recent years to understand its mechanisms, strategies, and potential solutions. This research examines essential factors that affect how online users behave regarding security in cyberspace, focusing on phishing attacks through the Health Belief Model (HBM). Understanding what influences users' security behaviors is crucial for building strong defenses. A survey was sent to students via WhatsApp and email, with 252 participants. The results were analyzed using quantitative methods. Principal Component Analysis (PCA) revealed that perceived barriers, self-efficacy, and privacy concerns were the main determinants of students' security behaviors. Students were particularly concerned about the misuse of their personal information. Despite varying levels of formal cybersecurity education, most students demonstrated confidence in configuring web browser security settings. The findings underscore the importance of tailored educational interventions and user-friendly security tools. Future research could explore additional security issues such as spyware, adware, and spam attacks. Additionally, leveraging machine learning and deep learning algorithms offers promising avenues for enhancing phishing detection and mitigation strategies. Furthermore, this study contributes to understanding cybersecurity behaviors, providing valuable insights for policymakers, educators, and developers to foster a safer online environment
Analyzing Perceptions of Maternal and Pediatric Care in Jakarta: An Integrated VADER and GloVe Analysis of Google Reviews in Mother and Child Hospitals
Get PDFIn the rapidly digitizing landscape of healthcare feedback, online reviews have become a vital source of patient-reported experiences. This study leverages sentiment analysis to decode the narrative content of Google reviews for Mother and Child Hospitals in Jakarta. Utilizing the VADER sentiment analysis tool and GloVe for keyword extraction, the research aimed to correlate qualitative sentiment with quantitative star ratings. This study meticulously processed and analyzed a selection of Google reviews using VADER for sentiment scoring and GloVe for refining the focus on relevant healthcare discussions. This methodological approach allowed for a comprehensive sentiment assessment of the reviews. The analysis revealed a prevalent positive sentiment in higher-rated reviews and negative sentiment in lower-rated reviews, with notable anomalies that underscore the complexity of patient experiences and perceptions. Specific aspects of care, including staff behavior, facility quality, and treatment efficacy, were recurrent themes in the feedback. These findings highlight the potential of patient-reported experiences in shaping healthcare practices and policy. The study emphasizes the importance of healthcare providers understanding and responding to patient feedback to improve care quality. Limitations such as the representativeness of online reviews and the challenges of sentiment analysis in capturing nuanced emotions are discussed. This study offers valuable insights into patient perceptions of maternal and pediatric care in Jakarta, affirming the significance of leveraging online reviews for healthcare quality monitoring and improvemen
A Low-Cost Nursing Robot with Telemedicine using ESP32 and Robot Operating System-based
Get PDFThe COVID-19 pandemic has presented unprecedented challenges to the healthcare sector, particularly frontline healthcare workers. These professionals face high infection risks and physical and mental exhaustion due to intensified workloads and staffing shortages. Robots are seen as a potential solution to this predicament, performing tasks such as delivering supplies and monitoring patients. However, widespread adoption of such robots, particularly in resource-constrained settings, has been hampered by the exorbitant costs associated with their acquisition and maintenance. To address this problem, the authors developed a low-cost nursing robot based on the ESP32 and the Robot Operating System (ROS). This robot facilitates hospital logistics and patient monitoring through telemedicine. The robot is controlled by remote control or Wi-Fi connection through the RViz Graphical User Interface (GUI) and uses odometry and PID control methods to follow specified paths autonomously. Accessible via local area networks and the Internet, the telemedicine system demonstrates robust performance with minimal X and Y axis control errors, zero packet loss, an average Round Trip Delay (RTD) of less than 150 ms, and jitter values of less than 20 ms, in line with TIPHON standards. This innovation provides a cost-effective solution to support healthcare workers during the ongoing health crisis. In future development, incorporating LiDAR, computer vision, and AI-based decision-making into the robot can facilitate obstacle detection and real-time decision-making to enable fully autonomous movement. These advancements will enhance the robot’s adaptability and accuracy in navigation and positioning
Comparative Performance Analysis of YOLO and Faster R-CNN in Detecting Species and Estimating the Weight of Grouper and Snapper Fish Using Digital Images
Get PDFGrouper and snapper fish are widely consumed species with high economic value in the global market. To determine their economic value, identifying the species and estimating the weight are essential in the pricing and quality determination of the traded fish. The commonly used manual methods are often time-consuming and labor-intensive. Based on this, a more effective computer-based method is needed for these repetitive tasks. This research aims to analyze the performance of two commonly used deep learning models, YOLO and Faster R-CNN, in detecting species and estimating the weight of specific grouper and snapper fish. The data used consisted of 2991 samples divided into 18 classes. This data was then augmented using rotate and flip features to create 6843 image samples. A threshold of 0.8 was used in the detection process, meaning objects detected with confidence below 0.8 would be ignored. Once trained, the performance of both models was tested using precision, recall, and accuracy parameters to assess how accurately the models predicted fish species from the input data and Mean Absolute Percentage Error (MAPE) to evaluate the estimation results of the models. There were differences in the quantitative evaluation results between the YOLO and Faster R-CNN models. The YOLO model achieved precision, recall, and accuracy rates of 0.98, 0.98, and 0.96, respectively, while the Faster R-CNN model had precision, recall, and accuracy rates of 0.97, 0.98, and 0.95, respectively. Additionally, the MAPE for weight estimation was 2.42% for image data and 3.66% for video data for the YOLO model. In contrast, for the Faster R-CNN model, the results were 14.62% for image data and 13.59% for video data. Thus, it can be concluded that the YOLO model provides better quantitative evaluation results compared to the Faster R-CNN model
Deep Learning-Based Early Dropout Prediction in University Online Learing
Get PDFWith the global transition of universities to online education due to COVID-19, the high dropout rate in online learning has become a critical challenge for higher education institutions. To address this issue, this study aims to develop a deep learning model for early dropout prediction in university online education. The proposed model was built by collecting and analyzing daily learning history data stored in the Learning Management System (LMS). Unlike previous studies that primarily relied on data collected at the end of the online learning period, this study analyzes students' behavioral data over time to more accurately identify students at risk. The research utilized data from a cyber university located in Seoul, South Korea, including approximately 30,574 student records and 12,014,610 learning history entries from the academic management system. To validate the model’s performance, data from the following academic year, which was not used for model training, was employed. The study compared the effectiveness of traditional machine learning methods with deep learning techniques (DNN and LSTM). Specifically, it proposed the LSTM-DNN model, which effectively learns both static learner information data and sequential learning history data. The results demonstrated that the LSTM-DNN model achieved a prediction accuracy of over 92%, confirming its effectiveness in providing real-time dropout risk assessments and predictive insights. Ultimately, this study proposes a novel approach to integrating real-time dropout prediction services into university Learning Management Systems (LMS), thereby contributing to student retention and academic success in online learning environments
Developing a Pre-Implementation Model for the ERP Material Management Module in the PVC Supply Industry in Indonesia
Get PDFRapid technological advancements have transformed the business landscape by introducing systems that enable more efficient and integrated operational management. One increasingly popular solution is Enterprise Resource Planning (ERP), which helps companies comprehensively manage business activities. However, many companies still face challenges in adopting ERP technology, particularly in managing inventory effectively and efficiently. One such company is PVC Supply Industry, which uses manual inventory recording. This research addresses the issues faced by the PVC Supply Industry as a case study to analyze pre-implementation ERP before transitioning to ERP implementation. The study begins with a literature review to identify 20 relevant ERP and inventory management indicators using the Semantic Scholar database. These 20 indicators were then included in a questionnaire distributed to individuals or groups with experience or understanding of ERP/IT. The responses were analyzed using the SmartPLS application, utilizing the IS success model, and categorizing the 20 indicators into four areas: people, process, technology, and ERP implementation readiness. The research results will provide four dominant indicators from these areas based on the SmartPLS analysis. The dominant indicators are P3 (human resource management), PC4 (material requirement planning), TC3 (prototype), and RE (ERP system implementation). These indicators will be applied to a Figma prototype using the prototyping method. The Figma prototype is designed to offer a rough user interface of the ERP material management module to address the issues encountered by the PVC Supply Industry as a case study
A Review of Technology Acceptance Model Application in User Acceptance of Autonomous Vehicles
Get PDFTouting the merits of reducing traffic fatalities by eliminating human errors and promoting social benefits, autonomous vehicle technology has been rapidly developed over the last few decades. Using the systematic review approach, this study provides an overview of current studies that apply the Technology Acceptance Model (TAM) in shaping user acceptance of autonomous vehicles. Based on a set of inclusion and exclusion criteria, a total of 16 articles out of 792,364 articles were retained for further analysis. The factors that have garnered the most attention from researchers are the technical and psychological factors, with the most frequent constructs integrated in these studies being perceived ease of use, followed by perceived usefulness, trust, attitude, perceived enjoyment, and perceived innovativeness. This study presents three key findings. The first shows that 36 potential antecedents influencing AV adoption were incorporated into TAM. Excluding the baseline model antecedents, the most studied factors were trust, personal innovativeness, and perceived enjoyment. Trust is widely recognized as a crucial factor in AV adoption and requires longitudinal research, as it is a dynamic element that evolves. The second finding concerns the various causal relationships between the constructs. The results showed mixed outcomes, which may be due to differences in levels of automation examined, geographical contexts, and socio-demographic factors. Third, the gaps identified in Section 4 can guide policymakers, researchers, and automakers in developing effective future strategies and research directions. Ultimately, a deeper understanding of public acceptance can facilitate the safe deployment of AVs on roads, leading to benefits such as improved traffic safety and increased sustainability for society and the economy.