JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
Enhancing Relational Database Efficiency through Algorithmic Query Tuning in Virtual Memory Systems
Get PDFThe rapid evolution of virtual memory-based relational database systems has significantly advanced data processing capabilities. However, the efficiency of these systems largely depends on query execution optimization, which can be enhanced through algorithmic query tuning techniques. This study investigates the impact of these techniques on enhancing query performance in virtual memory-based relational databases. Various algorithmic methods were analyzed to optimize query execution plans, with a focus on key performance indicators such as execution time, CPU and memory usage, disk I/O, and cache hit ratio. The systematic application of these methods revealed effective strategies for performance enhancement. Results show substantial improvements in execution time, resource utilization, and scalability. This work offers valuable insights for database administrators and system architects, highlighting the role of algorithmic query tuning in managing the growing demands for data processing. Future research endeavors should explore the realm of AI-driven automation, with a particular focus on enhancing query optimization techniques. Additionally, there is a pressing need to investigate advanced security measures that safeguard data integrity within expansive, large-scale systems. By adopting innovative approaches, we can ensure robust protection and efficient performance in an increasingly data-driven world
Improved Hybrid of Artificial Gorilla Troops Optimizer and Simulated Annealing Algorithm to Address Global Optimization Problems
Get PDFMetaheuristics represent a category of optimization methods that surpass traditional exact methods by effectively and efficiently solving complex optimization problems. Feature subset selection (FS) is essential and important for addressing classification problems involving choosing optimal feature subsets based on specific criteria; this process is employed to decrease dimensionality and eliminate noise from datasets, which leads to simplicity of rules, aids data visualization, accelerates learning and enhances predictive accuracy. Among these algorithms, the artificial Gorilla Troop Optimization (GTO) is a recent metaheuristic algorithm miming gorillas' social behaviors. This study introduces a novel hybridized optimization approach to be tailored for feature subset selection. The suggested GTO-SA method combines the strengths of the artificial Gorilla Troop Optimization and simulated annealing (SA) to effectively explore and exploit the search space, aiming to achieve the best promising solutions. Simulated annealing was employed as an internal function to enhance exploitation capabilities that occasionally accept inferior solutions, which gives rise to refining the search process. The empirical evaluations were conducted based on sixteen medical datasets from the UCI repository. The results demonstrate that GTO-SA outperforms native GTO, as well as other methods, including Particle Swarm Optimization (PSO), Sine Cosine Algorithm (SCA), and Ant Lion Optimizer (ALO). The experiment results clearly showed the hybrid algorithm's efficiency in exploiting and exploring feature spaces to identify the finest features. Moreover, it significantly improved classification accuracy with minimal run time across all data sets utilized
Comparison of Machine Learning as an Inference Engine to Improve Expert Systems in Dengue Disease
Get PDFDengue disease remains a significant public health challenge in tropical and subtropical regions, with rising incidence and mortality rates over the past few decades. While expert systems have been developed for early detection, traditional approaches often rely on rigid rule-based inference engines, which are limited by their dependence on expert-defined structures and lack adaptability to evolving knowledge sources. This study introduces a novel approach to enhance the flexibility and adaptability of expert systems by integrating machine learning (ML) techniques into the inference engine, leveraging the growing availability of medical record data as a dynamic knowledge source. Using a dataset of 90 medical records, balanced to 126 items via the Synthetic Minority Over-sampling Technique (SMOTE), we evaluated the performance of multiple ML algorithms, including Decision Trees (DT), Support Vector Machines (SVM), and Artificial Neural Networks (ANN), against traditional models like Naive Bayes (NB) and K-Nearest Neighbors (KNN). The DT, SVM, and ANN models demonstrated exceptional performance, achieving average accuracy, precision, recall, and F1 scores of 97.73%, 98.33%, 97.22%, and 97.41%, respectively. The key innovation of this research lies in developing an adaptive inference engine that can dynamically learn from medical data, reducing reliance on static rule bases and enabling the expert system to evolve with new knowledge. This approach improves diagnostic accuracy and provides a scalable and flexible framework for addressing other infectious diseases. By bridging the gap between expert systems and machine learning, this study paves the way for more intelligent, data-driven healthcare solutions with significant implications for public health and disease management
Online Counseling on Global Issues: Systematic Literature Review
Get PDFThe integration of expertise in counseling with a deep comprehension of contemporary technology is essential. Developing a sustained method is crucial for creating a practical framework to address the psychological ramifications associated with the escalating complexities of global challenges. Therefore, this study was conducted to explore the use and challenges of online counseling (e-counseling) for global issues using the systematic literature review (SLR) method. The search was carried out in the Scopus database to obtain 637 documents after limitations in the year of publication, starting in 2020–2023. Another limitation was the use of the English language, and after quality assessment, a 25-article document analysis was conducted. The results showed that e-counseling was critical in addressing challenges and impacted many individuals in different regions. According to NVivo analysis, the practical implementation of online counseling (e-counseling) encountered several challenges, such as using potentially vulnerable technology, constraints within interpersonal relationships, and incorporating different methods
Entity Extraction in Indonesian Online News Using Named Entity Recognition (NER) with Hybrid Method Transformer, Word2Vec, Attention and Bi-LSTM
Get PDFNamed Entity Recognition (NER) is a crucial task in Natural Language Processing (NLP) that identifies entities such as person names, locations, and organizations within the text. While many NER studies have concentrated on the English language, there is a significant need for further research on Indonesian NER. Indonesia presents unique challenges due to its structural complexities, polysemy, and ambiguities. Conventional machine learning and deep learning techniques have been widely applied in NER; however, more detailed exploration into integrating these methods for performance improvement is needed. This study introduces a novel hybrid model, TWBiL, which combines Transformer mechanisms, Word2Vec embeddings, Bidirectional Long Short-Term Memory (Bi-LSTM), and Attention mechanisms to enhance NER performance on Indonesian text. TWBiL harnesses the strengths of each component to generate superior word vector representations, extract intricate sentence features, and disambiguate entities contextually. Our experimental results demonstrate the effectiveness of the proposed hybrid model, revealing a significant improvement in NER performance. Specifically, TWBiL achieves an F1-Score of 85.11 on an Indonesian online news dataset, outperforming the traditional Bi-LSTM model, which achieved a score of 75.18. The results indicate that TWBiL effectively reduces ambiguity and captures context more accurately, enhancing entity recognition. Future research should priorities reducing computational time when handling larger datasets without compromising overall NER performance. This study underscores the potential of integrating advanced deep learning techniques to tackle the unique challenges of Indonesian NER, thus providing a solid foundation for further advancements in the field
Context-Aware Job Recommender System
Get PDFContext-aware recommendation systems have emerged as essential to interactive web content and online job search. Primarily, since so many job offers are published on different online platforms, it can make the users take some time to find good opportunities that match exactly what they are looking for, as well as countless qualified candidates and other characteristics within that context, such as temporality. This comes as no surprise, as many practitioners and researchers have resorted to machine learning to create context-aware job recommendation systems that cater not only to job seekers. In this comparative paper, we have analyzed various machine-learning models for job recommendation systems. Four fundamental pillars are considered: accuracy, scalability, interpretability, and computational efficiency. This paper also studies the extent to which these models are contextual (e.g., how well they can model factors due to user preferences, job requirements, location, industry evolution, and temporality.) and can be used as a recommendation system. This study uses real-world employment data from actual employment statistics (through fixed-period analysis), professional networking platforms, and online job market platforms. The study does so purposefully to be comprehensive because it believes the lessons from remote work are generalizable. Still, the data is from a wide variety of job sectors, job positions, and locations. The group created a test environment for constructing and testing machine learning algorithms. Collaborative filtering, content-based, matrix factorization, deep learning, and many other hybrid approaches have obtained better results. This study was performed on Python with sci-kit-learn, pandas, and NumPy. The proposed system is a context-aware job recommender system that employs many machine learning algorithms to personalize job recommendations concerning user preferences and contextual information such as job location, industry status, and temporal dynamics. The findings underscore the importance of choosing machine learning models that are well-suited for job recommendation systems on a case-by-case basis. This comparative study intends to add to the art by providing algorithmic proof and practical advice to properly leverage machine learning models proposed in a naturalistic, messy setting of context-aware job recommendation systems.
Ship Trajectory Prediction Based on Spatial-temporal Data Using Long Short-Term Memory
Get PDFThe frequent exploitation of shipping lines by passengers increased traffic and exposed it to more significant dangers. Precise predictions for ship trajectory conditions at sea must be available to ensure safe navigation across the oceans. This article presents a trajectory prediction approach based on Long Short-Term Memory (LSTM) neural networks applied to time series Automatic Identification System (AIS) position data, expressed in spatial-temporal form. LSTM is highly suitable for ship trajectory predictions as it can capture long-term dependencies and spatial-temporal patterns existing in AIS data, since LSTM is targeted toward sequential data. The proposed model extracts ship trajectories from AIS data and utilizes an LSTM (Long Short-Term Memory) model to predict future ship movements based on historical patterns. The experiments demonstrate that it is effective in predicting where ships to navigate next, providing a valuable tool for enhancing traffic flow and improving navigation safety. The model with LSTM unit 500, tested on 3,478 ship trajectories, showed a median RMSE prediction error ranging from 0.0720 to 0.0841, with prediction M=8 coordinate a head having the highest error (0.0841) and M=2 and M=9 having the lowest (0.0720); the interquartile range (IQR) spanned from 0.0571 to 0.1006, and M=2 had the most outliers (302) while M=8 had the fewest (171), indicating varying prediction stability across different points. Despite these results, challenges remain in maintaining prediction stability across all points. Further optimization could enhance the model's performance and address these limitations by incorporating more complex spatial-temporal features or hybrid techniques
Design of Automatic Irrigation System For Post-Mining Land Reclamation
Get PDFpost-mining land reclamation poses a challenge in restoring degraded land's ecological function and productivity, requiring optimal rehabilitation to make it productive and environmentally friendly. A key challenge in reclamation is the availability of efficient water sources to support the revegetation process. Conventional irrigation systems are inefficient and require intensive monitoring. Therefore, an innovative solution in the form of an automatic irrigation system is needed to optimize water use and support sustainable plant growth. This study aims to design and develop a technology-based automatic irrigation system that combines soil moisture sensors, water pumps, sprinklers, solar panels, solenoid valves, and microcontrollers to regulate irrigation efficiently and on time. The methodology includes hardware and software design, integration of soil moisture sensors, a microcontroller as the control unit, and system field testing. The system is designed to activate irrigation based on real-time soil moisture levels automatically, ensuring water is only applied when needed. The system is expected to reduce excess water use and improve irrigation effectiveness across large and diverse areas. Results show that this automatic irrigation system can reduce water consumption by 34.2% compared to conventional methods. In addition, farmers can remotely manage irrigation via the Internet or mobile apps, reducing irrigation time by 75 minutes. This system holds the potential to be an innovative and sustainable solution for post-mining land reclamation, ushering in a new era of efficient and sustainable agriculture
Comparative Analysis of Human Detection using Depth Data and RGB Data with Kalman Filter: A Study on Haar and LBP Methods
Get PDFAccurate human detection in video streams with occlusions, illumination variances, and varying distances is crucial for various applications, including surveillance, human-computer interaction, and robotics. This study investigates the performance of two widely used object detection features, Haar-like and Local Binary Pattern (LBP), for detecting human upper bodies in color and depth images. The algorithms are combined with Adaptive Boosting Cascade classifiers to leverage the discriminative power of Haar-like features and LBP texture features. Extensive experiments were conducted on a dataset comprising color images and depth data captured from a Kinect camera to evaluate the algorithms' performance in terms of precision, recall, accuracy, F1-score, and computational efficiency measured in frames per second (fps). The results indicate that when tested on color images, the Haar-Cascade method outperforms LBP-Cascade, achieving higher precision (27.4% vs. 7.8%), recall (49.2% vs. 7.8%), accuracy (21.4% vs. 4.1%), and F1-score (35.2% vs. 7.8%), while maintaining a comparable computational speed (19.07 fps vs. 19.26 fps). However, when applied to depth data, the Haar-Cascade method, coupled with Kalman filtering, demonstrates significantly improved performance, achieving precision (79.3%), recall (79.3%), accuracy (65.8%), and F1-score (79.3%) above 70%, with a computational time of approximately 19.07 fps. The integration of Kalman filtering enhances the robustness and tracking capabilities of the system, making it a promising approach for real-world applications in human detection and monitoring. The findings suggest that depth information provides valuable cues for accurate human detection, enabling the Haar-Cascade algorithm to overcome challenges faced in color image analysis.
An Improved Okta-Net Convolutional Neural Network Framework for Automatic Batik Image Classification
Get PDFBatik is one of Indonesia's most important cultural arts and has received recognition from UNESCO. Batik has high artistic and historical value with a variety of patterns. Currently, Indonesia has 5,849 batik motifs which are generally classified based on shape, color, motif and symbolic meaning. The diversity of batik motifs makes it difficult for ordinary people to fully recognize them. This paper intends to develop an automatic framework for classifying batik motifs as a solution to overcome this issue. To develop this classification automation framework, the paper proposes a new architecture based on deep learning, which is named Okta-net. The architecture consists of 8 convolutional layers with separate convolution operations (SeparableConv2D). The output of the last convolution block will be fed to the fully connected layer using global average pooling. Meanwhile, in developing a deep learning model to classify batik image patterns, a dataset of 5 batik classes (motifs) was organized, consisting of 4,284 batik images. Through a series of experiments carried out, the proposed Okta-Net architecture succeeded in achieving satisfactory results with a validation accuracy of 93.17%, Precision of 91.60%, Recall of 92.28%, F-1 Score of 91.54%, and a loss of just 0.12%. Thus, it can be concluded that Okta-Net architecture can help preserve Indonesia's batik cultural heritage by accurate batik motif’s classification. Apart from that, based on a comparison of research outcomes, Okta-Net outperformed most of earlier studies, the majority of which had an accuracy of below 90%