JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
An Integrated Depok Smart City Evaluation
Get PDFGiven the complicated pressures brought on by the fast pace of urbanization, innovative and long-lasting solutions to the issues arising from urban expansion are needed. To ensure a greater standard of life for their citizens and make sustainable growth one of their long-term goals, cities will need to make more inventive, persistent, and successful changes to their infrastructure. Nonetheless, smart cities require complex solutions to problems involving ICT, economics, government, social issues, the environment, and transportation. The sustainability of smart cities is now a topic that academics, environmental policymakers, and governmental organizations are more interested in. Depok's smart city must be evaluated to determine its capacity to fulfill the desired vision to help implement the Movement Towards 100 Smart Cities. This study offers an evaluation approach for the Depok smart city. Three indices were used to construct an integrated evaluation approach: the IMD Smart City Index 2023, The Cities of the Future Index, and the Global Power City Index. None of the indexes' results include all six of the Depok Smart City's necessary dimensions. Thus, the advice was to merge the three indices into an integrated evaluation approach for evaluating the six primary dimensions of the Depok Smart City. The results of this study also offer a sample measurement statement according to Depok Smart City. Furthermore, follow-up actions that the government or stakeholders can take to improve Depok's smart city performance include implementing the integrated matrix indicators and evaluating their validity and relevance in the real world.
Utilization of WebGIS for Visualization of the Distribution of Tourist Destination Religious Objects in Nagari Batuhampar of Lima Puluh Kota Regency, West Sumatera Province
Get PDFNagari Batuhampar has several tourist attractions planned as object tourist destinations in the strategic plan. However, publication and presentation on social media are less effective in promoting the unique characteristics of tourist attractions. This research aims to identify the distribution of tourist destination objects in Nagari Batuhampar, followed by comprehensive information. The type of research used is descriptive survey research with the waterfall method, which consists of requirement analysis, system analysis, system implementation, system testing, system evaluation, operation, and maintenance. Data collecting techniques include observation using GPS and documentation, interviews to obtain information for web development, and questionnaires. Furthermore, the built-in data application QGIS 3.32.3” Lima” is open source. WebGIS, built using the Database Management System (DBMS) approach, is designed as software to manage big data. Big data is meant to be a collection of lots of data tailored to the project being carried out, such as mapping the distribution of public facilities and village potential. In this research, DBMS focuses on spatial data and religious and supporting tourism attributes. This is focused on data on religious and supporting tourism attributes. The result found that historical religious tourist attractions dominated the distribution of attractions in Nagari Batuhampar. The WebGIS of Tourist Destination Object was constructed using a waterfall method that was effectively created. This development was conducted through system evaluation tests, resulting in most respondents being satisfied with the process's performance.
A Novel Network Optimization Framework Based on Software-Defined Networking (SDN) and Deep Learning (DL) Approach
Get PDFAccess to networks and the Internet has multiplied, and data traffic is growing exponentially and quickly. High network utilization, along with varied traffic types in the network, poses a considerable challenge and impact on the ICT Infrastructure, particularly affecting the performance and responsiveness of real-time application users who will experience slowness and poor performance. Conventional/traditional Quality of Service (QoS) mechanisms, designed to ensure reliable and efficient data transmission, are increasingly insufficient due to their static nature and inability to adapt to the dynamic demands of modern networks. As such, this study introduces a Novel Network Optimization Framework leveraging the combined strengths of Software-Defined Networking (SDN) and Deep Learning (DL) to dynamically manage multiple QoS of network devices in enterprise and campus network environments. The proposed system is a dynamic QoS that utilizes SDN's global monitoring and centralized management control capabilities to programmatically control network devices, ensuring that sensitive traffic is allocated with appropriate bandwidth and minimized latency. Concurrently, DL algorithms enhance the framework's decision-making process by proposing an accurate preferred configuration for the best adequate bandwidth for sensitive traffic transmission. This integration facilitates real-time adjustments to network conditions and improves overall network performance by ensuring high-priority applications receive the bandwidth they require without manual/human intervention. By providing a dynamic, intelligent solution to QoS management, this framework represents a significant step forward in developing more adaptable, resilient, and efficient networks capable of supporting the demands of contemporary and future digital ecosystems
Comparison of Convolutional Neural Networks Transfer Learning Models for Disease Classification of Food Crop
Get PDFIndonesia is an agricultural country with 29% of the workforce working in the agricultural sector, however, farmers' knowledge and practices depend on informal local wisdom based on inherited past practices. Moreover, identifying diseases in plants is difficult to do with human vision so that intelligent technology is needed. In this paper, an architecture of CNN models such as MobileNetV2, ResNetV50, InceptionV3 and DenseNet121 will be built to detect diseases based on leaf images of several crops obtained from the agroai dataset containing multiple crops namely bean, chili, corn, potato, tomato and tea. The model is used through transfer learning for feature extraction of the trained model with imagenet weights, with 4 fully connected layers. Each model for each crop will be compared to get the best model based on the accuracy of training, evaluation and testing. ResNet50 has the best performance for four type of plants, including bean plants with training accuracy of 99.49%, validation of 99.52%, testing of 98.96%, chili plants with training accuracy of 98.03%, evaluation of 98.75%, testing of 100%, tea plants with training accuracy of 99.62%, evaluation of 99.6%, testing of 99.74% and tomato plants with training accuracy of 99.62%, validation of 99.7%, testing of 99.37%. Moreover, MobileNetV3 has the best performance for 2 types of crops that is corn with training accuracy of 99.22%, validation of 99.69%, testing of 99.55%, and potato with training accuracy of 99.62%, evaluation of 99.60%, testing of 99.74%
The Effects of Imbalanced Datasets on Machine Learning Algorithms in Predicting Student Performance
Get PDFPredictive analytics technologies are becoming increasingly popular in higher education institutions. Students' grades are one of the most critical performance indicators educators can use to predict their academic achievement. Academics have developed numerous techniques and machine-learning approaches for predicting student grades over the last several decades. Although much work has been done, a practical model is still lacking, mainly when dealing with imbalanced datasets. This study examines the impact of imbalanced datasets on machine learning models' accuracy and reliability in predicting student performance. This study compares the performance of two popular machine learning algorithms, Logistic Regression and Random Forest, in predicting student grades. Secondly, the study examines the impact of imbalanced datasets on these algorithms' performance metrics and generalization capabilities. Results indicate that the Random Forest (RF) algorithm, with an accuracy of 98%, outperforms Logistic Regression (LR), which achieved 91% accuracy. Furthermore, the performance of both models is significantly impacted by imbalanced datasets. In particular, LR struggles to accurately predict minor classes, while RF also faces difficulties, though to a lesser extent. Addressing class imbalance is crucial, notably affecting model bias and prediction accuracy. This is especially important for higher education institutes aiming to enhance the accuracy of student grade predictions, emphasizing the need for balanced datasets to achieve robust predictive models
Automated UML Class Diagram Generation from Textual Requirements Using NLP Techniques
Get PDFTranslating textual requirements into precise Unified Modeling Language (UML) class diagrams poses challenges due to the unstructured and often ambiguous nature of text, which can lead to inconsistencies and misunderstandings during the initial stages of software development. Current methods often struggle with effectively addressing these challenges due to limitations in handling diverse and complex textual requirements, which may result in incomplete or inaccurate UML diagrams. This study aims to propose a Natural Language Processing (NLP) model that analyzes and comprehends textual requirements to extract relevant information for generating UML class diagrams, ensuring accuracy and consistency between the diagrams and requirement descriptions. The research employs a four-step approach: preprocessing to handle text noise and redundancy, sentence classification to distinguish between "class" and "relationship" sentences, syntactic analysis to examine grammatical structures, and UML class diagram generation based on predefined rules. The results show that the model achieved a classification accuracy of 88.46% with a high Area Under the Curve (AUC) value of 0.9287, indicating robust performance in distinguishing between class definitions and relationships. This study highlights that existing methods may not fully address the nuances of translating complex textual requirements into accurate UML diagrams. This study successfully demonstrates an automated method for generating UML class diagrams from textual requirements and suggests that future research could expand datasets, optimize feature extraction, explore advanced models, and develop automated rule generation methods for further improvements
An Innovative Approach for Improving Navigation Performance of Robust Land-Based Vehicles
Get PDFThe Extended Kalman Filter (EKF) stands as a prominent choice within navigation systems, particularly in scenarios involving the integration of a Reduced Inertial Sensor System (RISS) with the Global Positioning System (GPS). However, despite its widespread adoption, the EKF grapples with many challenges, including the propensity to underestimate filter uncertainties, contend with unreliable GPS signals, and confront errors stemming from linearization processes. These issues invariably contribute to a decline in overall system performance. Considering these challenges, this paper endeavors to introduce a groundbreaking integration algorithm to mitigate the inherent shortcomings of the EKF. The proposed algorithm employs innovative strategies to address these challenges comprehensively. Firstly, it incorporates a dynamic self-tuning mechanism meticulously designed to improve filter configuration in real-time, ensuring adaptability to varying operating conditions. The algorithm also integrates a meticulously engineered GPS Integrity algorithm to filter out mistaken readings and bolster the reliability of the navigation solution. Furthermore, the algorithm adopts the Unscented Kalman Filter (UKF), renowned for handling non-linearities directly, thereby cutting the need for the cumbersome linearization procedures inherent in the EKF. Comparative evaluations against the traditional EKF method prove the effectiveness of the proposed approach. Significant performance enhancements are evident using two datasets from a VTI SCC1300-D04 IMU unit compared to high-precision Novatel SPAN ground truth data. These improvements are quantified through RMSE analysis, showing substantial strides in navigation accuracy. Overall, the results underscore the transformative potential of the proposed integration algorithm in advancing navigation system capabilities
Application Development Model E-Commerce in Traditional Markets Using the TOGAF Framework
Get PDFThe existence of traditional markets is increasingly being eroded by the presence of online market models that already utilize various e-commerce platforms. However, the TOGAF Framework provides a ray of hope. Even though the existence of traditional markets is unlikely to disappear, if traders in traditional markets do not try to adapt to using e-commerce, it will have an impact on losing customers, losing turnover, and will even affect the acceleration of their business. Architectural design for system development in traditional market activities can provide an important picture of how to advance traditional market industries that will not be out of competition with modern markets. TOGAF (The Open Group Architecture Framework) provides a detailed method of how patterns build, manage, and implement enterprise architecture and information systems called the Architecture Development Method. Analysis using the TOGAF Framework provides an overview of how to plan, design, develop and implement information system architectures for traditional markets. The results of this design analysis are an information technology architectural design which is the basis for traditional market managers, especially in Indonesia, in advancing the business processes of traders in the market so that they are not unable to compete with modern markets but with gap analysis constraints that must really be considered with regard to internet infrastructure problems in the market, sellers who are not fully able to use technology due to age and not all forms of goods on the market can be sold through e-commerce platforms. This gap analysis will make e-commerce implementation slow to implement
UI/UX Redesign of SH-UPI App Using Design Thinking Framework
Get PDFThe rise of smart home technology has significantly impacted people’s behavior and lifestyle, especially when controlling household electronics remotely. Smart homes have become increasingly commoditized in the last decades, resulting in several vendors offering various commercial products and their variants. Universitas Pendidikan Indonesia (UPI) has created the smart home platform “SH-UPI,” which includes the Smart LED Bulb RGB. The platform’s mobile app can be downloaded from the Play Store or the official website (www.sh-upi.com). The SH-UPI is one of Indonesia's leading smart homes that adopted Internet-of-Things (IoT). However, customers have reported issues with the app’s user interface (UI) and user experience (UX), prompting a redesign to address these concerns and stay competitive. This study employs the design thinking method of empathizing, defining, imagining, prototyping, and testing. The define stage to prototype stage resulted in a newly designed prototype for the SH-UPI app. Testing involved evaluating the prototype using metrics like the User Experience Questionnaire (UEQ) and System Usability Scale (SUS). The test results showed an increase in the UEQ parameter value, exceeding the initial average scores of -0.8 to 0.8. Additionally, there was an improvement compared to benchmark scores, which initially ranged from below average to poor but now range from above average to excellent. The SUS score also improved, rising from 59.75 (grade D) to 83.625 (grade A). The findings of this study can be a valuable resource for developing SH-UPI apps with significantly enhanced UI/UX
Enhancing Early Detection of Melanoma: A Deep Learning Approach for Skin Cancer Prediction
Get PDFMelanoma, a form of skin cancer, is a substantial global public health threat due to its rising prevalence and the potential for severe outcomes if not promptly identified and managed. Detecting skin cancer lesions in their first stages enhances patient outcomes and decreases mortality rates. The core issue investigated in this research paper is the enduring problem of early skin cancer prediction. In the past, individuals often lacked awareness of their skin cancer condition until it had reached late stages. Consequently, this resulted in delayed diagnoses, which restricted the available treatment options and perhaps led to worse outcomes. This research focuses on finding key attributes and methods in a specialized dataset to effectively differentiate between benign and potentially malignant skin lesions, particularly the implementation of an early-stage skin cancer prediction model. It aims to accurately categorize skin mole pictures as benign or malignant using a Convolutional Neural Network (CNN) model built within the PyTorch framework. The primary aim of this study was to enhance the accuracy and effectiveness of diagnosing skin problems by implementing deep learning algorithms to automate the process of showing such conditions. The model underwent training using 3600 skin mole images sourced from the ISIC-Archive on a GPU RTX 3080. Its outstanding performance is shown by an F1 score of 0.8496 and an accuracy rate of 85%. This research aims to create a predictive model and offer a practical solution that healthcare professionals can readily use for early skin cancer prediction