cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Jumanto
Contact Email
jumanto@mail.unnes.ac.id
Phone
+628164243462
Journal Mail Official
sji@mail.unnes.ac.id
Editorial Address
Ruang 114 Gedung D2 Lamtai 1, Jurusan Ilmu Komputer Universitas Negeri Semarang, Indonesia
Location
Kota semarang,
Jawa tengah
INDONESIA
Scientific Journal of Informatics
Published by Universitas Negeri Semarang
ISSN : 24077658     EISSN : 24600040     DOI : https://doi.org/10.15294/sji.vxxix.xxxx
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering
Scientific Journal of Informatics (p-ISSN 2407-7658 | e-ISSN 2460-0040) published by the Department of Computer Science, Universitas Negeri Semarang, a scientific journal of Information Systems and Information Technology which includes scholarly writings on pure research and applied research in the field of information systems and information technology as well as a review-general review of the development of the theory, methods, and related applied sciences. The SJI publishes 4 issues in a calendar year (February, May, August, November).
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 25 Documents
 1   2   3 
Search results for , issue "Vol. 11 No. 4: November 2024" : 25 Documents clear
Pneumothorax Detection System in Thoracic Radiography Images Using CNN Method Fardana, Nouvel Izza; Isnanto, R. Rizal; Nurhayati, Oky Dwi
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.16635

Abstract

Purpose: This research aims to develop an automatic pneumothorax detection system using Convolutional Neural Networks (CNN) to classify thoracic radiography images. By leveraging CNN's effectiveness in identifying medical abnormalities, the system seeks to enhance diagnostic accuracy, reduce evaluation time, and minimize subjective interpretation errors. The output will provide a predicted label of "pneumothorax" or "non-pneumothorax," facilitating faster clinical treatment and improving diagnostic services while supporting radiologists in making more accurate and efficient decisions for this critical condition. Methods: This research employs an experimental deep learning approach using Convolutional Neural Networks (CNN) to detect pneumothorax in thoracic radiography images. The CNN model is trained on an annotated dataset with preprocessing steps, including zooming, brightness adjustment, flipping and format adjustment, followed by performance evaluation using accuracy, precision, recall, and F1 score metrics. Result: The results showed that the CNN model detected pneumothorax with 79.59% accuracy, a loss of 1.3056, and 1,092 correct predictions out of 1,372 test data. Precision was 51.12%, recall 78.62%, and F1 score 61.96%, confirming the system's potential, though further optimization is needed. Novelty: The novelty of this research lies in developing an automated pneumothorax detection system using a CNN architecture, improving diagnostic accuracy and efficiency. Despite high accuracy, precision and recall can be improved. Future research can focus on optimizing the model and applying data augmentation techniques.
Implementation of K-Nearest Neighbor in Case-Based Reasoning for Mental Health Diagnosis Systems Pamungkas, Ardian; Isnanto , R Rizal; Nugraheni , Dinar Mutiara Kusumo
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.19912

Abstract

Purpose: Assessing a model that employs the K-Nearest Neighbor (KNN) technique within Case-Based Reasoning (CBR) for diagnosing mental health disorders, concentrating on conditions such as anxiety, depression, stress, and normalcy, while enhancing its efficacy through the utilization of historical case data for more accurate and tailored diagnostic suggestions. Methods: This study implements the KNN method in CBR to create a mental health diagnosis system that can provide accurate results without the need for complex models or intensive training. This method effectively addresses various patient needs by utilizing previous case data to provide a personalized and case-based diagnosis. This system is designed to tackle mental health issues like anxiety, depression, and academic stress, utilizing a case study of students from ITBK Bukit Pengharapan. Result: This study developed a KNN-based model for mental health diagnosis, achieving 84.62% accuracy on test data. Data processing techniques like text mining, oversampling, and cosine similarity improved performance. With an optimal K value of 2, the model achieved 88% precision, 85% recall, and an F1-score of 84%. The anxiety label performed perfectly, with 100% precision, recall, and F1-score. Novelty: This study adds innovation by integrating the rarely used CBR and KNN algorithms for mental health diagnosis systems. Innovative techniques like text mining, oversampling to get around data integration, and cosine similarity computations, which greatly enhance model performance, assist this strategy. Because this method improves accuracy and expedites the diagnosis process, both of which support clinical decision-making, it may be able to help mental health professionals.
Improve Software Defect Prediction using Particle Swarm Optimization and Synthetic Minority Over-sampling Technique Amirullah, Afif; Umi Laili Yuhana; Muhammad Alfian
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.16808

Abstract

Purpose: Early detection of software defects is essential to prevent problems with software maintenance. Although much machine learning research has been used to predict software defects, most have not paid attention to the problems of data imbalance and feature correlation. This research focuses on overcoming the problems of imbalance dataset. It provides new insights into the impact of these two feature extraction techniques in improving the accuracy of software defect prediction. Methods: This research compares three algorithms: Random Forest, Logistic Regression, and XGBoost, with the application of PSO for feature selection and SMOTE to overcome the problem of imbalanced data. Comparison of algorithm performance is measured using F1-Score, Precision, Recall, and Accuracy metrics to evaluate the effectiveness of each approach. Result: This research demonstrates the potential of SMOTE and PSO techniques in enhancing the performance of software defect detection models, particularly in ensemble algorithms like Random Forest (RF) and XGBoost (XGB). The application of SMOTE and PSO resulted in a significant increase in RF accuracy to 87.63%, XGB to 85.40%, but a decrease in Logistic Regression (LR) accuracy to 72.98%. The F1-Score, Precision, and Recall metrics showed substantial improvements in RF and XGB, but not in LR due to the decrease in accuracy, highlighting the impact of the research findings. Novelty: Based on the comparison results, it is proven that the SMOTE and PSO algorithms can improve the Random Forest and XGB models for predicting software defect.
Sentiment Analysis on SocialMedia Using TF-IDF Vectorization and H2O Gradient Boosting for Student Anxiety Detection Ningsih, Maylinna Rahayu; Unjung, Jumanto
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v12i1.20582

Abstract

Purpose: Mental health issues are now a concern for many people. Anxiety or often called Anxiety that is excessive and prolonged has also become the forefront of various psychological disorders that trigger impacts such as stress to suicide. People using social media platforms tend to be a medium for expressing opinions sharing information and even expressing daily emotions. Many studies have shown a correlation between expressing emotional statements on social media and mental disorders. This research aims to conduct sentiment analysis of Anxiety on social media using H2O Gradient Boosting by implementing TF-IDF Vectorization which is set to max feature. Methods: This research utilizes 6980 post data from social media. The method applied is by conducting Exploratory Data Analysis then Data preprocessing, Tf-Idf Vectoriztion with max feature experiments 100, 250, 500, 1000 and 2000, H2O Gradient Boosting Model, Cross Validation, and Model performance evaluation. Result: The results of this study show good model performance through max feature TF-IDF = 250 with an accuracy value of 99%, Specificity 99.57%, and Eror Rate of 0.0106. Novelty: So that the use of the H2O Gradient Boosting model succeeded in providing good performance in classifying anxiety sentiment.
Hybrid Quantum Representation and Hilbert Scrambling for Robust Image Watermarking Sari, Christy Atika; Abdussalam, Abdussalam; Rachmawanto, Eko Hari; Islam, Hussain Md Mehedul
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.10140

Abstract

Purpose: This work aims to apply Quantum Hilbert Scrambling to enhance the security and integrity of image watermarking without affecting visual quality degradation. Further conception of the surveyed methods could result in a very good solution to conventional methods of watermarking in solving some problems of digital image security and integrity with new concepts of quantum computing. Methods: The paper reviews Quantum Hilbert Scrambling, whose computational complexity is . The process involves encoding the image into a quantum state, permuting qubits by the Hilbert curve, and embedding a watermark using quantum gates. Result: The quantitative performance evaluation metrics, like Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM), have shown high Peak Signal to Noise Ratio (PSNR) values from 56.13 dB to 57.87 dB and Structural Similarity Index (SSIM) from 0.9985 to 0.9990, correspondingly. This justifies the fact that the quality degradation is very slight and the fine details of the structure are well maintained. Novelty: The proposed method uniquely integrates quantum computing with traditional watermarking steps for a secure and effective approach in digital watermarking. Further development should focus on improving the quantum circuit regarding computation efficiency, extending the applicability of the method to a wide range of images, and various situations in watermarking, and finding hybrid approaches by combining quantum and classical approaches towards better performance and scalability.
Comparison of Digital Forensic Tools for Drug Trafficking Cases on Instagram Messenger using NIST Method Nahdli, Muhammad Fahmi Mubarok; Riadi, Imam; Biddinika, Muhammad Kunta
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13463

Abstract

Purpose: Cybercrime is an unlawful act that utilizes computer technology and the development of the internet. Cybercrime can occur on all electronic devices, including Android smartphones. Forensic handling, particularly mobile forensics, has become crucial in addressing drug trafficking cases conducted through Instagram. As the primary device for accessing Instagram, smartphones store digital data that can serve as evidence in investigations. This research aims to produce a more accurate comparison of results in analyzing Instagram Messenger data containing content related to drug trafficking. Methods: The digital evidence data used in this research included five types of data: text chat, account, image, audio, and image view once. The forensic tools for obtaining digital evidence were MOBILedit, Belkasoft, Mobile Forensic SPF, and Magnet Axiom. The method proposed in this research followed the NIST framework, which consists of four stages: collection, examination, analysis, and reporting. This research followed the NIST framework because it is widely recognized in the field of digital forensics and provides a comprehensive guideline for handling digital evidence. Result: Research results showed that Magnet Axiom had the best performance in digital forensic analysis, with a success rate of 74.1%. MOBILedit Forensic had a success rate of 62.5%, indicating lower performance. Mobile Forensic SPF had a success rate of 44.6%. In comparison, Belkasoft had the lowest success rate of 23.2%, showing that this software could be more effective in detecting and analyzing digital data than the others. Novelty: In this study, the analysis process was conducted using four digital forensic tools, each showing variations in terms of efficiency and effectiveness. Each tool has advantages and disadvantages regarding speed, accuracy, and ability to extract and manage data.
Comparison of KNN and CNN Algorithms for Gender Classification Based on Eye Images Wicaksono, Rizky Dwi; Fajar Shidiq, Guruh
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13529

Abstract

Purpose: This study explores gender classification using iris images and compares two methods k-nearest neighbors (KNN) and convolutional neural networks (CNN). Most research has focused on facial recognition. However, iris classification is more unique and accurate. This research addresses a gap in gender classification using iris images. It also tests the effectiveness of CNN and KNN for this task. Methods: This study used 11,525 iris images from Kaggle. Of these, 6,323 were male and 5,202 were female. The authors split the data into training (75%) and testing (25%). Preprocessing involved normalizing and augmenting images by rotating, scaling, shifting, and reflecting the them. Pixel values were also adjusted. The study compared the KNN algorithm, using Euclidean distance and 16 neighbors, with a CNN model. The CNN had layers for convolution, pooling, and density. The authors performed evaluation using accuracy, precision, recall, F1-score, and confusion matrix. Result: The KNN model demonstrated 81% accuracy. It identified males with 87% precision but only 70% recall. Meanwhile, the CNN model was better, achieving 93% accuracy with 94% precision and 95% recall for males. The CNN model outperformed KNN for females in precision, recall, and F1-score, indicating its superior ability to learn patterns and classify gender from iris images. Novelty: CNN outperforms KNN in classifying gender from iris images. It effectively recognizes patterns and achieves high accuracy. The study shows CNN’s superiority in biometric tasks, suggesting that future research should balance datasets and test better models, as well as combining models for better performance.
Music Genre Classification Using Mel Frequency Cepstral Coefficients and Artificial Neural Networks: A Novel Approach Alamsyah, Alamsyah; Ardiansyah, Fahmi; Kholiq, Abdul
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13660

Abstract

Purpose: Music is an artistic expression with many categories in various genres and styles, characterized by its melodic and harmonic compositions. Music genre classification is crucial because genres serve as descriptors commonly used to organize large music collections, especially on the internet and in widely used applications like JOOX and Spotify. The aim of this research is to implement the Mel Frequency Cepstral Coefficients (MFCC) feature extraction method to generate numerical features from a set of specific music tracks. This collection of information will then be classified using machine learning. Methods: The method used in this study begins with combining the "GTZAN Dataset - Music Genre Classification" with additional data from TikTok and YouTube. The total dataset consists of 1,200 audio files, divided into 12 classes. The MFCC extraction process generates numerical representations of acoustic characteristics, which are then processed using Artificial Neural Networks. Result: The experiments demonstrate that increasing the amount of data is crucial, as it can enhance both variation and accuracy. The average accuracy achieved in this study is 91.42%, while the highest accuracy reaches 92.16%. These findings indicate that this study outperforms previous studies. Novelty: The novelty of this research lies in the integration of dynamic social media data (TikTok and YouTube) to enrich the standard GTZAN dataset, the repetition of the MFCC feature extraction process, and the combination of MFCC with Artificial Neural Networks (ANN).
Evaluation of User Experience in the Banjarbaru Disdukcapil Public Service Application Using User Experience Questionnaire and System Usability Scale Martalisa, Asri; Wahyu Saputro, Setyo; Turianto Nugrahadi, Dodon; Abadi, Friska; Budiman, Irwan
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13780

Abstract

Purpose: Dukcapil Banjarbaru is an online-based government agency application used for various public services. According to the complaint report from Disdukcapil Banjarbaru, several users have reported similar problems and difficulties. The application has received a rating of 3.3 stars from approximately 24.000 users on the Google Play Store. Therefore, researchers conducted a user experience analysis using the UEQ methods and a usability evaluation using the SUS methods. Methods: This research analyzes user experience in applications using the UEQ to identify issues faced by users and evaluate usability through the System Usability Scale. The UEQ method is chosen for its efficiency and simplicity in assessing user experience within an application. The SUS method is employed because it is an effective approach for obtaining reliable statistical data and generating clear and accurate scores. Result: The UEQ benchmark results show that the scales for Attractiveness (1.59), Efficiency (1.68), Accuracy (1.66), and Stimulation (1.54) are categorized as "Good." The scales for clarity (1.37) and novelty (0.80) are classified as "Above Average." Meanwhile, the SUS score of 65 positions the application within the "acceptable" category for the acceptability range, the "D" category on the grade scale, and the "OK" category for adjective ratings. This indicates that while the Banjarbaru Dukcapil application has good usability, it requires improvements based on the total SUS score, which reveals several critical areas with scores below the average (258.4). Novelty: In this research, solutions for improvements are provided to Disdukcapil based on each aspect to improve the quality of the application, thereby offering better services to users.
Comparative Analysis of CNN Architectures in Siamese Networks with Test-Time Augmentation for Trademark Image Similarity Detection Suyahman; Sunardi; Murinto
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13811

Abstract

Purpose: This study aims to enhance the detection of trademark image similarity by conducting a comparative analysis of various Convolutional Neural Network (CNN) architectures within Siamese networks, integrated with test-time augmentation techniques. Existing methods often face challenges in accurately capturing subtle visual similarities between trademarks due to limitations in feature extraction and generalization capabilities. The research seeks to identify the most effective CNN architecture for this task and to assess the impact of test-time augmentation on model performance. Methods: The study implements Siamese networks utilizing three distinct CNN architectures: VGG16, VGG19, and ResNet50. Each network is trained on a dataset of trademark images to learn deep feature representations that can discriminate between similar and dissimilar trademarks. During the evaluation phase, test-time augmentation (TTA) is applied to enhance model robustness by averaging predictions over multiple augmented versions of the input images. TTA includes transformations such as random rotations (up to 40%), width and height shifts (up to 20%), random shear transformations, zooming (up to 20%), horizontal and vertical flips, and random brightness adjustments. Result: Experimental findings reveal that the Siamese network based on VGG19 achieves the highest accuracy at 98.82%, outperforming the VGG16-based network with an accuracy of 97.07% and the ResNet50-based network with 50.00% accuracy. The application of TTA has improved performance across all models, with the VGG19 model receiving the highest improvement. The extremely low accuracy of ResNet50 can be attributed to its misinterpretation of original trademark images as close-forged ones, probably due to overfitting or lack of an efficient ability in generalizing very fine visual features. Novelty: The study conducted a comparative analysis of CNN architectures, namely VGG16, VGG19, and ResNet50 in Siamese networks for trademark image similarity detection.

Page 2 of 3 | Total Record : 25

 1   2   3 

Filter by Year

2024 2024


Reset
Filter By Issues
All Issue Vol. 13 No. 1: February 2026 Vol. 12 No. 4: November 2025 Vol. 12 No. 3: August 2025 Vol. 12 No. 2: May 2025 Vol. 12 No. 1: February 2025 Vol. 11 No. 4: November 2024 Vol. 11 No. 3: August 2024 Vol. 11 No. 2: May 2024