JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
Visualization of Accounting and Indigenous People Research: A Bibliometric Review Using R
Full text linkThis study aims to map out the evolution of research trends on accounting and Indigenous peoples by using bibliometric analysis. Most bibliometric literature articles rely on basic graphical representations generated by computer systems. The methodology for conducting bibliometric analysis presented in this paper consists of three stages, namely data collection, software selection and analysis. This study used published papers from the Scopus database was carried out on 13 June 2024 and found 42 indexed research publications on the topic of accounting and indigenous people between 1999 and 2023. The map of research development in the field of accounting and Indigenous people is obtained through the export process, which was analyzed using the R Biblioshiny application program. The findings demonstrated a development trend with a static increase in the number of publications about accounting and research on Indigenous people. Besides, the results show that the journal with the most publication and impact is the Accounting, Auditing, and Accountability Journal. The country with the most objects of study is Australia. The development of research related to accounting and Indigenous People was growing, although not too massive. Along with these conditions, various trends in Accounting and Indigenous People Research topics grew. The results of this study also indicate that the most widely used topic keywords are Accounting, Indigenous, People, and Research. The findings of this study provide scholars with a comprehensive understanding of the current research work in the field of accounting and indigenous people and its future directions
Understanding Search Behavior in the Simulated Kalman Filter Algorithm
Full text linkIn computational optimization, metaheuristic algorithms are crucial for solving complex and dynamic problems. It is important to fully understand how an algorithm searches, as it helps to improve the algorithm and its applications in various domains. This paper provides a detailed analysis of how the Simulated Kalman Filter algorithm searches for optimal solutions. The SKF algorithm is an optimization method inspired by the Kalman filter estimation techniques. The algorithm was introduced in 2015 to address unimodal problems. Since its inception, the SKF algorithm has undergone improvements and is used to solve a range of optimization problems. Our study aims to bridge the gap in existing research by investigating how SKF effectively balances the search space exploration and known solution exploitation. Through systematic experimentation using the Brown function as a benchmark, we explored the social dynamics and movement style of the SKF algorithm, in addition to the convergence efficiency and accuracy. When we applied the same approach as suggested in the referenced paper, we gained insights into SKF’s unique strengths and limitations of SKF when compared to other algorithms. The findings illustrate SKF’s unique capabilities in handling the exploration-exploitation trade-off. This study helps to set the foundation for creating more advanced algorithms and optimization strategies in the future. Future research will examine how enhancements to the SKF algorithm impact and enhance its search behavior
Challenges of Agile Software Development in the Banking Sector: A Systematic Literature Review
Full text linkThe banking industry is expected to thrive, generate profits, and contribute to national development and societal welfare. However, this sector is susceptible to volatility caused by global and domestic economic fluctuations. This research aims to identify and address challenges related explicitly to implementing agile methodologies within the banking sector. The study utilized a Systematic Literature Review (SLR) approach based on the guidelines provided by Kitchenham. A substantial number of academic journals (1,933) were analyzed during this review. Among the vast pool of literature, 28 relevant studies were extracted. These studies were chosen because they provided insights into the challenges of implementing agile practices in the banking domain. The analysis and categorization of these studies were structured according to the Project Management Body of Knowledge (PMBOK) 6th edition framework. This framework was employed to organize and understand the identified challenges systematically. The study's primary finding is that the most prevalent challenge encountered in the context of agile development within the banking sector is "Project Resource Management." In essence, effectively managing and allocating resources is a significant hurdle banks face when adopting agile methodologies. The challenges related to resource management are not confined to a single aspect. Instead, they encompass various dimensions, including human resources, technological resources, and organizational factors. This suggests that challenges in agile banking are multifaceted, involving issues related to people, technology, and the structure and processes within banking organizations
Analysis of Student Perceptions on Blended Learning Using Learning Management System (LMS) for Physical Education, Sports, and Health Courses
Full text linkThis study investigates student perceptions of LMS-based Blended Learning in Physical Education, Sports, and Health subjects at Public Junior High School 25 in Barru Regency, South Sulawesi, Indonesia. A descriptive quantitative design was utilized for this research. Probability sampling was employed to ensure representativeness. Data was collected through a structured questionnaire consisting of twenty- five items designed to measure four key aspects of LMS- based blended learning: e- learning knowledge, e- learning accessibility, e- learning usefulness, and e- learning usage satisfaction. The reliability of the questionnaire was confirmed via Cronbach's α, which produced a value of 0 830, and McDonald's ω, yielding a value of 0 0.850, indicating strong internal consistency and reliability of the instrument. Results showed that 82. 55% of respondents agreed or strongly agreed that e- learning knowledge is vital for supporting blended learning, suggesting awareness and confidence among students regarding the role of digital learning tools in enhancing their educational experiences. Additionally, 61. 61.41% agreed or strongly agreed that e- learning accessibility significantly aids the implementation of blended learning, emphasizing that easy access to LMS platforms is crucial for student engagement. Furthermore, 60. 16% acknowledged the importance of e- learning usefulness in the current educational landscape, highlighting a widespread recognition of digital tools' significance in education. Lastly, 53. 83% stated satisfaction with e- learning usage is a key factor influencing successful blended learning experiences. These findings indicate a favorable perception among students toward LMS-based blended learning in physical education, sports, and health subjects. The study emphasizes the importance of e- learning knowledge, accessibility, usefulness, and satisfaction for creating effective blended learning environments. Further research is suggested to examine the long-term effects of LMS-based blended learning on student outcomes across diverse educational settings
A Deep Learning Approach Using VGG16 to Classify Beef and Pork Images
Full text linkThere are 87.2% of the Muslim population in Indonesia, which makes Indonesia one of the countries with the largest Muslim population in the world. As a Muslim, it is supposed to carry out and stay away from the commands that Allah SWT commands, one of which is in QS. Al-maidah:3, one of the commands in the verse is not to consume haram food such as pork. Even so, it turns out that many traders in Indonesia still cheat to get more significant profits, namely by counterfeiting beef and pork. The lack of public knowledge supports this situation to differentiate between the two types of meat. Therefore, the classification process is used to distinguish the two kinds of meat using the convolutional neural network approach with VGG16 with several preprocessing stages. Two primary stages are used during the preprocessing stage: scaling and contrast enhancement. The VGG16 algorithm gets very good results by getting an accuracy value of 99.6% of the test results using 4,500 images for training data and 500 images for testing data. To compare the effectiveness of these techniques, it is recommended to use alternative CNN architectures, such as mobilNet, ResNet, and GoogleNet. More investigation is also required to gather more varied datasets, enabling the ultimate goal to achieve the best possible categorization, even when using cell phone cameras or with dim or fuzzy photos
Predicting the Next Day's Closing Price of Stock Indices Using Machine Learning and Deep Learning Algorithms
Full text linkShare prices are a critical factor in a stock index’s worth but are never constant. Thus, an effective method of predicting share prices is needed. This is where machine learning comes in. This research discusses the applicability of machine learning algorithms, precisely long short-term memory, artificial neural networks, and linear regression in predicting share prices. Additionally, this research goes in-depth, explaining how each algorithm functions. These three algorithms were implemented using the financial dataset of the S&P 500, one of the more known stock indices out there. Data was collected from Yahoo Finance for 34 years, from 1990 to 2023. Then, the algorithms mentioned were used to train a model using the collected dataset. All three algorithms were measured using three performance metrics: mean absolute error, R-squared score, and mean absolute percentage error. The final implementation involved training them by only using 1-day lagged features to create a model that can predict the next day's closing price. All the algorithms performed considerably well, with linear regression being the best, followed by artificial neural networks and long short-term memory being the worst. Finally, the implemented algorithms were used to predict the closing prices of other stock indices, NASDAQ and Hang Seng Index. All algorithms performed well and followed the same trend, wherein linear regression performed the best and long- and short-term memory the worst. Future research should be conducted to explore the possibilities of utilizing lagged features along with external features like GDP growth rate, political trends, etc
Performance Improvement of Cosine Similarity Algorithm with Bidirectional Encoder Representations from Transformers on Abstract Document Similarity Detection
Full text linkIn thesis courses or final projects, students are required to be able to conduct research by the science they are engaged in, find innovations, solve problems, and foster a culture and critical mindset. However, the issue that is often encountered is plagiarism. Plagiarism is taking a work that can be in the form of someone else's opinion and making it seem as if it is your own. The step in applying technology that can be done is to carry out early detection of the similarity of documents written by students. In this case, the document that will be detected is an abstract that must be collected by students when submitting a thesis title. The algorithm used is a cosine similarity algorithm, which is computationally efficient because of its ease of interpretation and compatibility with large-scale data. This research was carried out using two schematic approaches: bidirectional encoder representations from transformers (BERT) and not bidirectional encoder representations from transformers (BERT). The corpus data used in this study was 1450 data of student thesis abstract documents, with the test using 10 data to see the performance of the cosine similarity algorithm in detecting the similarity of abstract documents. The results showed that documents with optimization using the Bidirectional Encoder Representations from Transformers (BERT) approach had better results, with an average performance improvement of 23.48%
Identification of Indonesian Rupiah Paper Currency Denominations Using First Order Statistical Feature Extraction and k-Nearest Neighbor
Full text linkThe rupiah currency is legal tender in Indonesia and issued by Bank Indonesia. The paper rupiah currency has undergone many changes. Although each paper rupiah has its characteristics, errors can occur when distinguishing the value of paper rupiah between denominations. With the rapid development of image processing that can be utilized in human life, the problem of errors in distinguishing the value of paper rupiah can be overcome with image processing techniques, where paper rupiah data can be identified using these techniques. This study focuses on the feature extraction and classification process. The dataset used in this study is the image of the paper rupiah with the 2022 emission. The image extraction process uses First-Order Statistical Characteristic Extraction to obtain the characteristics of each image object. In addition, K-Nearest Neighbor (KNN) is used to classify paper rupiah denominations. The accuracy, sensitivity, and specificity values are calculated to indicate the level of success of the test data compared to the training data. The testing process is carried out with k values of 1, 3, 5, and 7 on the front and back sides of the currency for all datasets. The highest accuracy value was obtained when k = 3. This test produced an average Accuracy value of 92.08%, an average Sensitivity value of 64.22%, and an average Specificity value of 96.61%. This research can be developed for more affordable counterfeit money detection, such as smartphone applications or portable devices that the general public can use
Optimization of Herbal Plant Classification Using Hybrid Method Particle Swarm Optimization With Support Vector Machine
Full text linkThe classification process applied in this study helps identify the many kinds of herbal plants. Herbal plant leaf features are used based on color, shape, and texture. Particle Swarm Optimization and Support Vector Machine (PSO-SVM) hybridization are applied in the classification process to increase classification and identification accuracy. A well-liked metaheuristic approach for solving optimization issues is Particle Swarm Optimization (PSO). Particles look around the search area for the best responses. A particle swarm is initially initialized randomly within the search area via the PSO algorithm. Every particle's mobility is determined by both its own experience and the experiences of the other particles in the swarm. Each particle keeps track of the best solution it has ever found and the swarm's most extraordinary remedy that has so far been discovered. The Hybrid approach concurrently selects features for the SVM and optimizes its parameters. The kernel function's gamma value non-linearly maps an input space to a high-dimensional feature space. At the same time, the C parameter determines the trade-off between fitting error minimization and model complexity. The Gaussian kernel parameter is set to determine the optimal parameter value of the RBF kernel function. Feature selection solves the issue by eliminating redundant, associated, and irrelevant features. A confusion matrix is utilized in the evaluation to gauge the system's performance. The results demonstrated an improvement in accuracy, with the hybrid PSO-SVM using test data achieving an accuracy of 98% compared to the SVM method, achieving a 91% accuracy
Optimising iCadet Assignment through User Profiling
Full text linkIndustry Cadetship programme is a programme that assigns penultimate year students to companies matching their profiles, bridging academic learning and industry skills. Manual data analysis for assignments is time-intensive, prompting this study’s objectives: (i) propose an algorithm to optimize student-company assignment by using the student and company profiles, (ii) propose a method for the assignment of lecturers to company, and (iii) use similarity measure techniques to recommend companies with similar characteristics. Data was collected from a university's student, company, and lecturer datasets. To assign students to companies, the Haversine, OpenStreetMap, and NetworkX were used to calculate the shortest geographical distance between the students and the companies; evaluated based on mean, variance, standard deviation, and utilization rate. For the lecturer assignment, cosine similarity was applied to measure the similarity between domain descriptions and company or lecturer information after performing Voyage AI embeddings. Lecturers are assigned to companies based on the highest domain similarity scores. The performance was evaluated using accuracy, precision, recall, and F1- score. Findings showed embedding techniques significantly enhanced the matching process, with accuracy improved from 0.464 to 0.6071, precision increased from 0.417 to 0.5058, recall saw an equal rise from 0.464 to 0.6071, and the F1-score advanced from 0.417 to 0.5264. Longer descriptive inputs further improved performance, with accuracy rising from 0.6154 to 0.7692, precision from 0.5744 to 0.7751, recall remaining steady at 0.7692, and F1-score increasing from 0.5807 to 0.7484. This work can be extended to explore job portal dataset by aligning profiles with geography and specialization