JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
UAV-Based Segmentation and Correlation Analysis of Vegetation Indices for Cassava Crop Health Assessment
Get PDFCassava, an essential staple food with diverse applications, has been relatively underexplored in terms of health analysis using vegetation indices. Conventional field surveys face challenges in covering large areas due to resource constraints. Recent advancements in remote monitoring techniques, such as satellite imagery and Unmanned Aerial Vehicles (UAVs), offer a promising alternative. While satellite imagery enables broad-scale surveys, its limited spatial resolution restricts detailed analyses of individual plants or smaller ecosystems. UAV-based vegetation surveys commonly utilize Vegetation Indices (VI) to assess unique spectral information. This study investigated UAV-based methods for mapping cassava distribution in the Telaga Kahuripan smallholder plantation in Bogor, Indonesia, focusing on UAV imagery, segmentation, and vegetation indices to evaluate cassava plant health at 2, 5, and 8 months of age. The results revealed significant variations in vegetation indices across different cassava plant ages. Particularly, the highest values observed at 5 months of age indicated substantial growth, with NDVI and GNDVI values exhibiting R2 ranging from 0.95 to 0.98, indicating a strong correlation. The robust correlation between NDVI and GNDVI implies that both indices can effectively predict plant health using UAV-based monitoring. Comparisons with existing studies suggest potential variations attributable to factors such as geographical location, environmental conditions, and cultivation practices. Understanding these variations is crucial for refining monitoring techniques and informing agricultural practices. Consequently, the findings have implications for enhancing cassava health monitoring and optimizing agricultural practices to ensure sustainable crop production
Improved Content-based Image Retrieval by Improving Low-Level Features Detection with Artificial Neural Networks
Get PDFWith the rapidly growing number of digital photos being taken using different devices in recent years, significant attention has been brought to improving the ability to match these images. However, the reliance on traditional Content-Based Image Retrieval (CBIR) techniques on certain features, e.g., objects, low-level features, or colors, in these images has caused a semantic gap in the matching results. Recent techniques rely on multiple features to reduce this gap and employ artificial neural networks (ANNs) to produce a single similarity measure that represents the overall similarity between two images. Additionally, several studies have suggested that these networks better detect low-level features when processing the input image in grayscale rather than separate color channels. In this study, we propose a new methodology that allows ANNs to process colored and grayscale versions of images simultaneously, producing a more accurate similarity measure by accurately considering the high-level, low-level, and colors in the input images. The model implementation is based on the Yolo V8 neural network architecture. It is evaluated against recent state-of-the-art methods using several datasets, including MIRFLICKR-25K, NUS-WIDE, MS-COCO, Pascal VOC2007, and Pascal VOC2012. We assess the model's performance using three well-established metrics: NDCG, ACG, and wMAP. The proposed technique outperformed all existing methods in terms of NDCG and wMAP. Experimental results demonstrate that this method has also achieved high-performance measures with significant improvements and more stable results at different datasets of different images and classes, especially when the quality of the results is measured using the NDCG. Such an improvement illustrates the importance of using the grayscale version of the image as an input to the neural network to improve its ability to recognize local features better than only providing the image in RGB
Chatbot Adoption Model in Determining Student Career Path Development: Pilot Study
Get PDFA career decision is incredibly essential in one's life. It shapes one's future role in society, influences professional development, and can lead to success and fulfillment. Making a sound and consistent career decision based on skills and interests is critical for personal and professional development. Since generative AI is an emerging and revolutionizing technology industry in the market, which is very good in generating contents, providing consultancies and answering questions in humanly fashion, integrating AI chatbots into the career planning process can help students to get more accurate and personalized advice for their future career. This pilot study emphasized the student’s adoption of chatbot technology for career selecting processes utilizing the extended Unified Theory of Acceptance and Use of Technology (UTAUT2) model with four additional constructs which influence the student’s career selection, namely: Perceived Student’s External Factors (PEF), Perceived Student’s Interest (PSN), Perceived Career Opportunities (PCO) and Perceived Self-Efficacy (PSF). An online survey was conducted, and 37 responses were received and analyzed. The measurement model produced a promising result, and the discriminant validity, construct reliability and validity of the model were confirmed with a Cronbach’s alpha (α) above 0.70 threshold and AVE over 0.5 cut-off for most of the constructs including the four above mentioned latent variables. However, the Price Value (PPV) and Facilitating Conditions (PFC) UTAUT2 constructs produced alpha () of 0.680 and 0.611 respectively which is still adequate since their AVE is above the 0.5 threshold. Consequently, their interpretation and conclusions should be approached with caution
Performance Evaluation of a Simple Feed-forward Deep Neural Network Model Applied to Annual Rainfall Anomaly Index (RAI) Over Indramayu, Indonesia
Get PDFIndramayu is a district in West Java that is known for being the leading producer of rice and brackish salt. The production of these two commodities is strongly influenced by hydroclimatological conditions, making accurate and reliable long-term estimates crucial. In this study, we evaluated a simple feed-forward deep neural network (DNN) model that could potentially be used as a candidate for statistical guidance to improve the accuracy of a mesoscale numerical climate model. We used the spatial average of the accumulated annual rainfall of the Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) data as an input time series with a time range from 1981 to 2022. This data was then processed into annual rainfall anomaly index (RAI) data. The Annual RAI was divided into training and test sets, and the feed-forward DNN model was fitted to the annual RAI in the training set. The accuracy of the model was then tested in the test set using the root-mean-square error (RMSE) metric. Our study shows that the feed-forward DNN model is unsuitable for estimating the annual RAI over Indramayu. The RMSE values are significantly high in the training and test sets
The Effect of Feature Selection on Machine Learning Classification
Get PDFHigh-dimensional datasets can lead to overfitting and computationally expensive model building on machine learning. This study uses a dimensionality reduction technique, namely feature selection techniques, to overcome these problems. Five feature selection methods were used, i.e., Chi-Square (CS), Information Gain (IG), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Least Absolute Shrinkage and Selection Operator (LASSO), and three classifier methods viz. Naïve Bayes, Extreme Gradient Boosting (XGB), and RF Classifier. The dataset used is the Heart Attack Analysis & Prediction Dataset. In this study, three scenarios of the best feature selection were carried out, namely: 1. selection of the best feature using a specific feature selection, 2. the intersection of selection of the best feature from the same category, 3. the intersection of selection of the best feature from the five proposed feature selection methods. The performance model is measured using accuracy, precision, recall, f1-score, AUC, and training time. This study reveals that feature selection is very effective in improving the performance of prediction models. Based on the experiment results, the best feature selection is CS and IG in the Filter Category with the XGB model. The best feature selected improved the performance of accuracy, precision, recall, f1-score, and AUC, i.e., 1.7%, 1%, 2.3%, 1.6%, and 0.2%, respectively. Meanwhile, training time requirements decreased by 23.5%. Feature selection with specific techniques performs better than feature selection by selecting the best features from the same category feature selection technique or various other feature selection methods.
Computational Visualization and Informatics Interaction Analysis of Daidzein Compound from Soybean (Glycine max L.) on Maltase-Glucoamylase Protein for Predictive Study of Intestinal Disaccharidase Deficiency
Get PDFThis study explores the potential of daidzein, a bioactive compound derived from soybean (Glycine max L.), as a maltase-glucoamylase protein inhibitor to address intestinal disaccharidase deficiency, utilizing in silico methodologies. The research supports Sustainable Development Goal 3: Good Health and Well-being by evaluating the binding interactions, physicochemical properties, and therapeutic potential of daidzein. Structural data of daidzein and maltase-glucoamylase were analyzed using PyMOL, PyRx, Protein Plus, and Lipinski’s Rule of Five to predict interaction mechanisms and drug-likeness. The methodological framework consisted of molecular docking and physicochemical analysis, including binding affinity and Root Mean Square Deviation (RMSD) evaluations. The docking results demonstrated strong and stable interactions between daidzein and the target protein, with binding affinities of -2.5 and -2.4 kcal/mol. Additionally, key physicochemical parameters—such as molecular weight (254) and log P (2.713)—indicated favorable drug-likeness and oral bioavailability. RMSD values supported the stability of daidzein within the enzyme’s active site. These findings suggest that daidzein may serve as a promising natural therapeutic agent for digestive disorders associated with enzyme deficiencies. The study also illustrates the efficiency of computational tools in the early stages of drug discovery, reducing reliance on laboratory testing. It is recommended that future research includes in vitro validations and preclinical studies to further assess the safety, efficacy, and pharmacokinetics of daidzein. Structural optimization to enhance target binding is also encouraged. Ultimately, this research contributes to the sustainable development of plant-based therapies for managing non-communicable diseases and improving digestive health
Bibliometric Analysis of AI-Based Prototype Proposal for User Security Awareness in Healthcare
Get PDFIn the realm of public healthcare, integrating information technology (IT) must be judiciously balanced with heightened security awareness among users, given the escalating frequency of cyberattacks targeting this sector. Despite the availability of various product and service solutions aimed at enhancing user security awareness, these efforts have yet to yield optimal outcomes. There is a pressing need for innovative approaches to bolster healthcare user security awareness through IT, particularly leveraging the rapidly advancing field of artificial intelligence (AI). This study conducts a comprehensive review of prior research on the application of AI, specifically Large Language Models (LLM), within the domain of healthcare cybersecurity from 2014 to 2024. The objective is to ascertain the volume of publications, trace the evolution of publication trends, and assess the potential and positioning of research in this area. Employing a bibliometric analysis methodology, this study analyzes a dataset comprising 1000 related publications indexed by Google Scholar. The findings reveal that publications concerning applying LLM AI in healthcare cybersecurity constituted 12.82% in 2023, with a significant increase to 87.18% in 2024, representing a 6.8-fold rise. The mapping of publication developments is categorized into 24 clusters, with large language models, healthcare, retrieval-augmented generation, LLM, artificial intelligence, and cybersecurity emerging as the six most frequently discussed keywords in the research landscape. Consequently, this study underscores the substantial potential for current and future research on the application of AI in healthcare cybersecurity, advocating for the development of AI-based solutions to enhance healthcare user security awareness
Integration of MQTT and Augmented Reality using Dobot Magician Lite Robot
Get PDFAs robotics education evolves, there is a growing demand for effective tools that offer hands-on learning experiences while addressing the traditional challenges such as limited interactivity, high costs, and the need for sophisticated technical knowledge. This paper explores the integration of wireless controllers and Augmented Reality (AR) image tracking with the Dobot Magician Lite robotic platform to enhance robotics learning. The proposed solution utilizes MQTT (Message Queuing Telemetry Transport), a lightweight messaging protocol, to enable intuitive and user-friendly wireless control of the robot. By using AR image tracking, students may see and operate virtual overlays that are aligned with the actual robot, bridging the gap between theoretical understanding and practical application. This technique not only streamlines the control procedure, but it also creates an interesting and immersive learning environment, making robotics education more accessible to a larger range of people. Team Alpha (n=4) and Team Beta (n=4) conducted testing in Indonesia and found promising results: average response time was 160 ms and 165 ms, respectively; movement accuracy was 1.5 mm and 1.7 mm; AR display quality received scores of 8.8 and 8.6; and user satisfaction ratings were 9.0 and 8.9. Both teams reported great system adaptability and minimal issue frequency, with Team Beta mentioning minor performance difficulties including occasional latency. These findings highlight the effectiveness and reliability of the proposed system, supporting its potential for broader application in robotics education
A New Feature Extraction Approach in Classification for Improving the Accuracy of Proteins
Get PDFProteins play a vital role in life as essential macromolecules, consisting of linear heteromeric biopolymers formed by amino acids covalently bonded through peptide bonds. They contribute to cell development and bolster the body's defense mechanisms. Post-translational modification processes, such as glycosylation, are necessary for proteins to function optimally. Glycosylation involves adding sugar groups to proteins, playing a critical role in various protein folding processes. Dysregulation of protein glycosylation can lead to diseases like Alzheimer's and cancer. Manual classification of glycosylated proteins is time-consuming, necessitating a faster approach. This study aims to expedite glycosylated protein classification using novel methods like AAindex, CTD, SABLE, hydrophobicity, and PseAAC for increased accuracy, comparing them with existing approaches. The dataset comprises protein sequences sourced from the openly accessible UniProt database. Results demonstrate that glycosylated protein prediction achieved 100% accuracy, surpassing previous approaches. Several features contributed to this improvement, with Hydrophobicity making a significant contribution at 24%, and PseAAC making the most significant contribution at 40% among the five extraction methods developed
Systematic Literature Review on Persuasive System Design Framework for Managing Curriculum Performance
Get PDFIntegrating digital resources into educational assessment has led to the widespread adoption of e-portfolios as tools for documenting and evaluating student achievement, thereby transforming traditional evaluation methods. However, the existing frameworks primarily focus on assessing academic performance, often neglecting the comprehensive monitoring of student’s co-curricular activities. To overcome current gaps in comprehensive student evaluation, this study introduces a conceptual framework incorporating persuasive system design (PSD) into an e-portfolio to facilitate efficient co-curricular performance monitoring in Malaysian secondary schools. To ensure a thorough approach to educational evaluation, it is essential to effectively monitor and manage academic and extracurricular performance to understand student progress comprehensively. By adding Physical Activity, Sports, and Co-curriculum Assessment (PAJSK) – specific categories and key PSD elements- primary task support, dialogue support, system credibility support, and social support- that are all designed to improve user engagement and system dependability in an educational environment, the framework builds on the Oinas-Kukkonen and Harijumaa PSD Model. This study adapts and discusses the persuasive design elements to meet the goals of educational assessment frameworks by comparing PSD implementation in e-health, e-tourism, e-commerce, and e-learning. The results offer an overview of developing a practical, engaging e-portfolio framework that facilitates comprehensive student evaluation, especially in educational environments focusing on co-curricular achievement