cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Putra Wanda
Contact Email
putra.wanda@respati.ac.id
Phone
+6287715730553
Journal Mail Official
ijicom@respati.ac.id
Editorial Address
Department of Informatics, University of Respati Yogyakarta
Location
Kab. sleman,
Daerah istimewa yogyakarta
INDONESIA
International Journal of Informatics and Computation
Published by Universitas Respati Yogyakarta
ISSN : 26858711     EISSN : 27145263     DOI : 10.35842/ijicom
Core Subject : Science,
Computer Science & IT
International Journal of Informatics and Computation (IJICOM) is an international, peer-reviewed, open-access journal, that publishes original theoretical and empirical work on the science of informatics and its application in multiple fields. Our concept of Informatics includes technologies of information and communication as well as the social, linguistic, and cultural changes that initiate, accompany, and complicate their development. IJICOM aims to be an international platform to exchange novel research results in simulation-based science across all scientific disciplines. It publishes advanced innovative, interdisciplinary research where complex multi-scale, multi-domain problems in science and engineering are solved, integrating sophisticated numerical methods, computation, data, networks, and novel devices. The scope of this journal includes IoT, 5G, Artificial Intelligence, sensor networks, and high-resolution imaging techniques. This new discipline in science combines computational thinking, modern computational methods, devices, and collateral technologies to address problems far beyond the scope of traditional numerical methods
Arjuna Subject : Ilmu Komputer (semua kategori) - Kecerdasan Buatan
Articles 61 Documents
 2   3   4   5   6 
IDS-GAN: Stepping up Intrusion Detection Method using GAN Algorithm Fan Haoyi; Sarah Anjani
International Journal of Informatics and Computation Vol. 5 No. 1 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i1.55

Abstract

Many computer network threats cause the security aspect to become the most critical problem. The intrusion detection system is a widely used practical security tool to prevent malicious traffic from penetrating networks and systems. To solve the issue, we construct a novel algorithm using Generative Adversarial Networks (GAN) to address the IDS security problem. In this paper, we propose an intrusion detection model using GAN by analyzing the extracted features of the network. To build our detection model, we collect the dataset, conduct pre-processing, train our model with several hyper-parameters to get the best accuracy, then test the model using the new data. Based on experimental results, the proposed model can produce a 0.00539 error rate and indicate a more accurate model to detect anomalies in the network traffic.
Robust Stock Price Prediction using Gated Recurrent Unit (GRU) Hamzah; Sugeng Winardi; Poly Endrayanto Eko Chrismawan; Rainbow Tambunan
International Journal of Informatics and Computation Vol. 5 No. 1 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i1.56

Abstract

Forecasting the direction of price movement of the stock market could yield significant profits. Traders use technical analysis, which is the study of price by scrutinizing past prices, to forecast the future price of the nickel stock price. Therefore, in this study, we propose Gated Recurrent Units (GRU) to predict nickel stock price trends. This research aims to produce an accurate nickel stock price trend prediction model. The research method utilized historical data on nickel stock prices from Yahoo Finance. The research results show that the model developed accurately predicted nickel stock price trends. From the RMSE, MAE, and MSE analysis results, the RMSE value was 0.0123, the MAE value was 0.0089, and the MSE value was 0.0002 on the test data.
FaceGAN: Robust Face Recognition using Generative Adversarial Networks (GAN) Algorithm Maryama Kurnia Amri; Bambang Sugiantoro
International Journal of Informatics and Computation Vol. 5 No. 1 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i1.57

Abstract

Generative Adversarial Networks (GANs) are a type of neural network that can generate synthetic images that are often indistinguishable from real ones. The article explores GAN to augment existing datasets or generate new ones for training classifiers. The competitive training process of GANs results in a generator network that can produce increasingly realistic images to create more diverse and balanced datasets for training classifiers. The article discusses several successful applications of GANs in image classification, including object recognition, face classification, and medical image analysis. The datasets used in this article are CelebA and FER2013. The CelebA dataset consists of 202,599 celebrity images with 40 attributes, such as gender, age, and facial hair. The FER2013 dataset consists of 35,887 images of faces with seven other emotions, including anger, disgust, fear, happiness, sadness, surprise, and neutral. The dataset is divided into training, validation, and test sets. We resized the images to 64x64 pixels and normalized the pixel values between -1 and 1, then trained a GAN model using the dataset. We evaluate the performance of our approach and compare it with several state-of-the-art methods, including Support Vector Machines (SVM) and Convolutional Neural Networks (CNN). We evaluate the performance of our approach and compare it with several state-of-the-art methods, including Support Vector Machines (SVM) and Convolutional Neural Networks (CNN), with the results that our approach outperforms SVM and CNN methods on both datasets, achieving a classification accuracy of 89.2% on CelebA and 72.5% in FER2013. Meanwhile, classification accuracy on SVM was 82.3% on CelebA and 65.4% on FER2013. Classification accuracy on CNN is 87.9% on CelebA and 70.8% on FER2013.
Comparative Approach for Intrusion Detection using CNN M. Aziz; Ahmad W
International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i2.58

Abstract

In the realm of computer network security, the role of intrusion detection is crucial for safeguarding systems against various threats. However, with the advancement of intrusion techniques, traditional detection methods have demonstrated constraints in recognizing novel attacks. This study tackles the urgent challenge of enhancing intrusion detection by employing Convolutional Neural Networks (CNN) algorithms, contrasting them with different machine learning methodologies like Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Gaussian Naive Bayes (GNB), Decision Trees, and Gradient Boosting (GBoost). The primary aim is to assess and compare the effectiveness of these algorithms utilizing an extensive dataset acquired from Kaggle, comprising 25,192 data entries and 42 attributes. Through the assessment of metrics such as accuracy, precision, recall, and F1-score, the findings reveal a nuanced profile of the strengths and weaknesses of each approach. Remarkably, CNN demonstrated remarkable accuracy, prompting further inquiry into its performance.
Enhancing Cardiovascular Diseases Classification using CNN Algorithm. Romanah H; Juwita Sampe Ruru
International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i2.60

Abstract

This study focuses on using machine learning algorithms to detect cardiovascular diseases, addressing the critical need for accurate and timely diagnosis of these conditions, which are significant contributors to global morbidity and mortality. The research aims to evaluate the performance of various machine learning algorithms such as Convolutional Neural Network (CNN), Support Vector Machine (SVM), Decision Tree, K-Nearest Neighbors (KNN), Gaussian Naive Bayes (GNB), and Gradient Boosting in categorizing patients into 'yes' or 'no' groups for cardiovascular diseases based on a thorough dataset. The methodology includes data preprocessing, feature selection, and model training and assessment. The results indicate that CNN and SVM demonstrate strong and balanced performance, whereas the Decision Tree shows high sensitivity but potential overfitting. These outcomes offer valuable insights for algorithm selection and model improvement in the detection of cardiovascular diseases, setting the groundwork for further research to enhance diagnostic accuracy, clinical relevance, and healthcare outcomes.
A Comparative Study of Detecting Twitter Spam using Deep Learning M Amman Said; Yaser Ahmad
International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i2.61

Abstract

This study addresses the escalating challenge of Twitter spam detection by leveraging the power of Convolutional Neural Networks (CNNs). With the proliferation of spam content on social media platforms, traditional machine learning algorithms have exhibited limitations in discerning intricate patterns within sequential data. The research problem centers on the need for a more robust and effective approach to distinguish spam tweets from legitimate content. The primary objective is to evaluate the performance of CNNs in comparison to baseline algorithms, including SVM, Decision Tree, KNN, Gaussian Naive Bayes, and Gradient Boosting. The research approach involves thorough data preprocessing, followed by model training and assessment using metrics like Confusion Matrix and Classification Report. The outcomes indicate that the CNN model outperforms the baseline algorithms, exhibiting superior levels of accuracy, precision, recall, and F1-score. These results highlight the promise of CNNs in reshaping the landscape of Twitter spam detection, presenting a more precise and effective approach to tackle the spread of spam content across social media platforms.This research contributes valuable insights for the development of advanced machine learning techniques in the domain of online security and spam detection.
Effective Ransomware Attacks Detection Using CNN Algorithm Huang J Jin; Cathern Hibbins
International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i2.62

Abstract

This study identified ransomware threats in social media platforms by evaluating the performance of Assessing different machine-learning algorithms in various aspects of detecting and classifying ransomware content. The primary problem revolves around the need to enhance cybersecurity within the dynamic landscape of social media, where users are increasingly susceptible to malicious attacks. The research objectives involve assessing the effectiveness of different algorithms, including Convolutional Neural Networks (CNN), Support Vector Machines (SVM), Decision Trees, K-Nearest Neighbors (KNN), Gaussian Naive Bayes (GNB), and Gradient Boosting (GBoost), in distinguishing between ransomware and benign content. A dataset consisting of 6,245 records with 15 features is employed to achieve this. The methods encompass data preprocessing, algorithm implementation, and performance evaluation using accuracy, precision, recall, and F1-score metrics. The research results revealed significant variations in algorithm performance, with Decision Tree and GBoost exhibiting exceptional accuracy while class imbalance challenges and model optimization issues were identified. These findings provide valuable insights into the complex realm of ransomware detection in social media, offering a foundation for future research and cybersecurity improvements in the digital space.
Improving Vehicle Detection in Challenging Datasets: YOLOv5s and Frozen Layers Analysis Ahmad Nanda Yuma Rafi; Mohamad Yusuf
International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v5i2.64

Abstract

Small datasets and imbalanced classes often cause problems when it used as primary research material. In case of classification and object detection, some researchers proposed Transfer Learning (TF) with several frozen layers. Moreover, YOLO (You Only Look Once) is one of the algorithms that works in real-time object detection. In this research, we focused on evaluating the YOLOv5s version of detecting vehicles in small and imbalanced datasets. The original YOLOv5s were trained and compared with YOLOv5s with freezing layers method (10 and 24 frozen layers). The experimental results of original YOLOv5s were precision score of 0.779, recall value of 0.933, mAP@0.5 of 0.93 and mAP@0.5:0.95 of 0.684 while YOLOv5s with 10 frozen layers where precision score was decreased to 0.639, but the other value increase with recall value of 0.939, mAP@0.5 of 0.951 and mAP@0.5:0.95 of 0.732. Overall, the version with 10 frozen layers demonstrated superior performance in addressing the challenges of small and imbalanced datasets, particularly excelling in recall and mAP metrics.
Enhancing Mental Health Disorders Classification using Convolutional Variational Autoencoder Sri Hasta Mulyani
International Journal of Informatics and Computation Vol. 6 No. 1 (2024): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v6i1.65

Abstract

This research investigates the application of Convolutional Variational Autoencoder (CVAE) for multi-class classification of mental health disorders. The study utilizes a diverse dataset comprising five classes: Normal, Anxiety, Depression, Loneliness, and Stress. The CVAE model effectively captures spatial dependencies and learns latent representations from the mental health disorder data. The classification results demonstrate high precision, recall, and F1 scores for all classes, indicating the model's robustness in distinguishing between different disorders accurately. The research contributes by leveraging the unique capabilities of CVAE, combining convolutional neural networks and variational autoencoders to enhance the accuracy and interpretability of the classification process. The findings highlight the potential of CVAE as a powerful tool for accurate and efficient mental health disorder classification. This research paves the way for further advancements in deep learning techniques, supporting improved diagnosis and personalized healthcare in mental health.
Enhancing Crime Prediction Using K-Means with PSO Ulumuddin
International Journal of Informatics and Computation Vol. 7 No. 1 (2025): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v7i1.68

Abstract

The effects of social media and modern approaches help offenders to achieve their crimes. This paper explores machine learning architecture to predict criminal crime cases by classifying each type of crime using K-Means which is optimized with PSO from the data the researcher got in the past mas. The clustering parameters use medium, light, and severe crime categories, each of them gets medium = 74, light = 46, and weight = 30. According to the experimental result, K-Means optimization with PSO can produce 0,12287 which uses SSE parameters while k-means performance gets results 0.885.

Page 4 of 7 | Total Record : 61

 2   3   4   5   6 

Filter by Year

2019 2025


Reset
Filter By Issues
All Issue Vol. 7 No. 1 (2025): International Journal of Informatics and Computation Vol. 6 No. 2 (2024): International Journal of Informatics and Computation Vol. 6 No. 1 (2024): International Journal of Informatics and Computation Vol. 5 No. 2 (2023): International Journal of Informatics and Computation Vol. 5 No. 1 (2023): International Journal of Informatics and Computation Vol. 4 No. 2 (2022): International Journal of Informatics and Computation Vol 3 No 2 (2021): International Journal of Informatics and Computation Vol 3 No 1 (2021): International Journal of Informatics and Computation Vol 2 No 2 (2020): International Journal of Informatics and Computation Vol 2 No 1 (2020): International Journal of Informatics and Computation Vol 1 No 2 (2019): International Journal of Informatics and Computation Vol.1, No.2, 2019 Vol 1 No 2 (2019): International Journal of Informatics and Computation Vol 1 No 1 (2019): International Journal of Informatics and Computation More Issue