cover
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
+6282161108110
Journal Mail Official
jurnal.josyc@gmail.com
Editorial Address
Jalan Sisingamangaraja No. 338, Medan, Sumatera Utara
Location
Kota medan,
Sumatera utara
INDONESIA
Journal of Computer System and Informatics (JoSYC)
ISSN : 27147150     EISSN : 27148912     DOI : -
Journal of Computer System and Informatics (JoSYC) covers the whole spectrum of Artificial Inteligent, Computer System, Informatics Technique which includes, but is not limited to: Soft Computing, Distributed Intelligent Systems, Database Management and Information Retrieval, Evolutionary computation and DNA/cellular/molecular computing, Fault detection, Green and Renewable Energy Systems, Human Interface, Human-Computer Interaction, Human Information Processing Hybrid and Distributed Algorithms, High Performance Computing, Information storage, Security, integrity, privacy and trust, Image and Speech Signal Processing, Knowledge Based Systems, Knowledge Networks, Multimedia and Applications, Networked Control Systems, Natural Language Processing Pattern Classification, Speech recognition and synthesis, Robotic Intelligence, Robustness Analysis, Social Intelligence, Ubiquitous, Grid and high performance computing, Virtual Reality in Engineering Applications Web and mobile Intelligence, Big Data
Articles 443 Documents
Machine Learning-Based GPS Spoofing Detection in UAV Networks: A Comparative Analysis of Anomaly Detection Models Airlangga, Gregorius
Journal of Computer System and Informatics (JoSYC) Vol 6 No 2 (2025): February 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i2.7033

Abstract

The increasing reliance on Global Positioning System (GPS) technology in Unmanned Aerial Vehicles (UAVs) has exposed them to cybersecurity threats, particularly GPS spoofing attacks that manipulate location data. This study explores the effectiveness of various machine learning-based approaches in detecting GPS spoofing in UAV communication networks. Supervised classification models, unsupervised anomaly detection techniques, and deep learning-based autoencoders are evaluated to determine their capability in identifying spoofed signals. The dataset used for training and testing contains multi-dimensional UAV network parameters with labeled GPS spoofing instances. Experimental results indicate that traditional anomaly detection models, such as Isolation Forest, One-Class SVM, and Local Outlier Factor, struggle with detection accuracy and exhibit high false-positive rates. The autoencoder-based approach achieves the highest accuracy (91.20%) but has poor precision (3.97%) and recall (4.73%), highlighting limitations in threshold selection and anomaly classification. Computational complexity analysis reveals that deep learning models, despite their accuracy advantages, require significant computational resources, making them less feasible for real-time UAV applications. This study identifies critical challenges in GPS spoofing detection, including dataset bias, environmental variability, and model hyperparameter sensitivity.
Deep Learning-Based Fetal Health Classification: A Comparative Analysis of Convolutional and Recurrent Neural Networks Airlangga, Gregorius
Journal of Computer System and Informatics (JoSYC) Vol 6 No 2 (2025): February 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i2.7056

Abstract

Fetal health monitoring plays a crucial role in prenatal care, enabling early detection of complications that may impact pregnancy outcomes. Traditional methods, including cardiotocography (CTG), rely on expert interpretation, which can introduce variability and potential misdiagnoses. In this study, deep learning techniques are employed to classify fetal health conditions based on CTG data. A comparative analysis is conducted on six architectures: Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional LSTM (BiLSTM), and Attention-based LSTM. The models are evaluated using accuracy, precision, recall, and F1-score under a 10-fold cross-validation framework. Results indicate that CNN outperforms all other models, achieving an accuracy of 97.18% due to its hierarchical feature extraction capabilities. GRU demonstrates competitive performance with an F1-score of 95.50% while maintaining computational efficiency. The study further includes a complexity analysis, revealing that recurrent models, particularly BiLSTM and Attention-LSTM, introduce significant computational overhead without yielding substantial performance gains. Potential threats to validity, including dataset bias and overfitting, are analyzed to ensure robust findings. The insights gained from this research highlight the advantages of CNN-based architectures in automated fetal health assessment and suggest future work integrating hybrid models and explainable AI techniques. These findings contribute to advancing AI-driven fetal monitoring systems, aiding clinical decision-making, and improving perinatal care.
Analisis Sentimen Komentar Pengguna Instagram Mengenai Pelaksanaan Pemilu 2024 dengan Naïve Bayes dan Lexicon-Based Dewi, Cahyani Rahma; Iskandar, Agus
Journal of Computer System and Informatics (JoSYC) Vol 6 No 1 (2024): November 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i1.5784

Abstract

The debate surrounding the implementation of the 2024 General Election has taken centre stage in Indonesia, especially on social media platforms that are favoured by the public. The change of leaders in Indonesia and the emotional differences that emerge in society are of significant concern. The search for leadership figures brings up various complex theoretical, conceptual, and cultural perspectives. This paper aims to analyse people's sentiment related to the 2024 general election by classifying sentiment as positive, negative, or neutral, aiding understanding of people's perceptions of candidates, relevant political issues, and voter behaviour patterns. The methodology involved collecting data using scrapping techniques from the social media platform Instagram using a combination of both Naïve Bayes Classifier and Lexicon-Based labelling algorithms. These two methods were used to conduct sentiment analysis towards the general election in this study. Sentiment analysis of the 2024 General Election using the Naive Bayes and InSet Lexicon models showed good results with an accuracy of 72% (precision negative 74%, neutral 54%, positive 70%; recall positive 62%, neutral 22%, negative 87%). This study successfully surpassed the accuracy of the previous model (72% accuracy, 70% precision, 72% recall) and revealed that negative sentiments were more prevalent in public opinion towards the 2024 General Election. This indicates that there is public dissatisfaction and doubt regarding the implementation of the election, which is thought to be triggered by technical problems and political uncertainty.
Investigasi File Carving pada Media Penyimpanan Menggunakan Framework Computer Forensic Investigative Process Fakhri, La Jupriadi; Riadi, Imam; Yudhana, Anton
Journal of Computer System and Informatics (JoSYC) Vol 6 No 1 (2024): November 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i1.6125

Abstract

One of the uses of digital storage media in the digital era that is still popular today is the use of flash drives as a means of transferring data between computer devices. Flash disks are often used as evidence in digital investigation cases. The risk of losing data is one of the main problems that society often faces. Data loss occurs for various reasons, such as user error, device failure, malware attack, or criminal acts such as hacking. The file carving technique is used to recover lost or deleted files from digital storage media with Foremost software. However, with so many file types, it is sometimes difficult to choose which file types to recover and how to ensure the authenticity of the files. This study uses the Computer Forensic Investigative Process (CFIP) Framework on a flash drive, which is used as evidence in a criminal case. Foremost software is used to perform file carving techniques on flash drives. The results showed that the data acquisition process using DC3DD succeeded in producing digital evidence with a hash value that is identical to the original file. Foremost software successfully recovered various file types, such as 9 image files with jpg file type, 5 audio files with mp3 file type, and 5 document files with pdf file type. Foremost shows a high success rate, with file carving accuracy of 90% for image files, and 62.5% for audio files and documents. The average success rate of Foremost software in returning evidence is 73.08%.
Prediksi Cuaca Menggunakan Data Historis dengan Algoritma Regresi Linear untuk Analisis Perubahan Suhu Pratama, Egi; Fatchan, Muhammad; Aguswin, Ahmad
Journal of Computer System and Informatics (JoSYC) Vol 6 No 2 (2025): February 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i2.6950

Abstract

Tokyo, the capital of Japan located on the state of Honshu, is facing with subtropical climate complexity, combining extreme temperature variations reached in hot summer season (>35 degrees) and cold winter season temperatures below 0 degrees. Current research explored the regression linear algorithm potential to predict daily maximum temperature within the context of complex urban weather dynamics. Based on the meteorology dataset collected in total of 639 days including key variables of temperature, humidity, rainfall, and air pressure, study developed weather prediction model. The outcomes demonstrated exceptional performance with Root Mean Squared Error at 0.80 and R-squared of 0.99, showing the near full coverage of model’s ability to capture all possible weather variability patterns. As a result, the research findings not only confirmed the effectiveness of linear regression for urban weather prediction but also open the possibility of similar model integration within more sophisticated weather forecast systems. Data-centered approach made significant contribution to the modern weather prediction technology responsive to urban society requirement.
Determining the Shortest Route for Eid Homecoming Route Using the Haversine Formula Method and A Star Algorithm Nurpandi, Finsa; Syarifah Sany, Diny
Journal of Computer System and Informatics (JoSYC) Vol 6 No 3 (2025): May 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i3.5911

Abstract

Eid homecoming, also known as "Mudik" in Indonesia, is an annual tradition that involves the mass movement of people from cities to their hometowns to celebrate the Eid al-Fitr holiday. Based on data from the Ministry of Transportation of the Republic of Indonesia, there are five major regions that serve as the primary destinations for homecoming travelers: West Java, Central Java, East Java, DI Yogyakarta, and the Jabodetabek region. The Central Java region is estimated to receive the largest number of homecoming travelers, with an estimated 61.6 million people. Given the potential for human movement of up to 193.6 million people during this period, it poses a significant traffic burden, and an efficient determination of the shortest route is crucial. To address this challenge, the study utilizes the Haversine Formula, which calculates the distance between two geographic points on the earth's surface by considering the curvature of the earth. This approach provides a more accurate distance estimate compared to traditional linear distance calculations. Additionally, the A* algorithm is employed to determine the shortest path from the starting point to the destination. The A* algorithm combines the distance calculation results from the Haversine Formula as a heuristic component in the search process, effectively optimizing the route selection. The results of the A* algorithm search identified the shortest route for homecoming travelers, which starts from the city of Jakarta, passes through West Karawang, Indramayu, Cirebon, Tegal, Pekalongan, Semarang, and Salatiga, before reaching the final destination of Klaten. This optimized route covers a total distance of 599.4 km, providing an efficient and cost-effective option for travelers during the Eid homecoming period
Model Deep Learning Berbasis Inception V3 untuk Klasifikasi Penyakit Daun Apel Menggunakan Citra Digital Arifin Nur, Khairun Nisa; Wanto, Anjar; Windarto, Agus Perdana; Solikhun, Solikhun
Journal of Computer System and Informatics (JoSYC) Vol 6 No 3 (2025): May 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i3.7003

Abstract

Apple plants have high economic value, but their productivity is often disrupted by leaf diseases that can reduce quality and yield. Apple leaf disease identification is still largely performed manually, which is prone to errors and requires specialized expertise. Therefore, a method is needed to improve the accuracy and efficiency of apple leaf disease classification. This study aims to enhance the accuracy of apple leaf disease classification by implementing the Convolutional Neural Network (CNN) architecture, specifically Inception V3. The method involves collecting images of infected apple leaves, data preprocessing, and model training and evaluation. The results show that the Inception V3 model achieved an accuracy of 96%, which is higher than previous methods. The main advantage of this architecture lies in its ability to capture features at multiple scales simultaneously, improving the model’s ability to recognize disease patterns more accurately. With these findings, this study contributes to the development of AI-based plant disease detection technology and provides a practical solution for farmers to enhance apple farming productivity.
Hybrid G2M Weighting and WASPAS Method for Business Partner Selection: A Decision Support Approach Wang, Junhai; Setiawansyah, Setiawansyah; Alita, Debby
Journal of Computer System and Informatics (JoSYC) Vol 6 No 3 (2025): May 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i3.7229

Abstract

Choosing the right business partner is a crucial factor in the success and continuity of a company's operations. The main issue in selecting business partners is the complexity of balancing various interconnected and often conflicting factors. Another problem lies in the subjectivity and limitations of information. Evaluators or decision-makers may have differing views on the priority of criteria or the interpretation of the available data. This study proposes a hybrid method-based decision support system approach that combines G2M Weighting and WASPAS to address the challenges in complex and uncertain multi-criteria evaluations. The G2M method is used to objectively determine the weight of criteria based on geometric averages in gray environments, so as to be able to capture data variability and uncertainty. Furthermore, the WASPAS method is applied to calculate the final value and rank the alternative business partners based on a combination of additive and multiplicative approaches. The ranking chart for business partner selection using the G2M Weighting and WASPAS method shows that Partner G gets the highest score of 9.93E+03, followed by Partner A and Partner E who have the same score of 9.43E+03. Meanwhile, Partner D had the lowest score, which was 5.97E+03. This ranking of business partner selection shows that Partner G is the best choice as a business partner based on the evaluation method used. The results of the study show that this hybrid approach provides more accurate, stable, and comprehensive evaluation results than conventional methods. This approach can be an effective solution for companies in supporting the strategic decision-making process in choosing the best business partners.
Implementation of Apriori Algorithms to Analyze and Determine Consumer Purchase Patterns in Gadget Stores as Sales Increase Strategy Simanullang, Rahma Yuni; ', Khairunnisa; Wanny, Puspita; Utari, Utari; Novelan, Muhammad Syahputra
Journal of Computer System and Informatics (JoSYC) Vol 6 No 3 (2025): May 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i3.7355

Abstract

This study aims to identify the pattern of product purchases that often occur simultaneously at a gadget store in order to develop a more effective sales strategy. The research problem focuses on how to find associations between products based on sales transaction data. The proposed solution is to apply data mining techniques, specifically a priori algorithms, to analyze transaction data and find significant association rules. The A priori algorithm is used through several stages, including the calculation of support for each item, the elimination of items with support below the minimum threshold, the formation of itemset combinations, and the calculation of confidence to generate association rules. The results showed two association rules that met the minimum confidence threshold (60%), namely: (1) If customers buy USB-C, they tend to buy Powerbank (confidence: 67%), and (2) If customers buy Smartwatches, they tend to buy Screen Protectors (confidence: 67%), and (3) If customers buy Screen Protectors, they tend to buy Smartwatches (confidence: 100%). These patterns can be used by the store for strategic product placement and bundling promotions.
Analisa Perbandingan Metode Trend Moment dan Regresi Linear dalam Prediksi Kurs Mata Uang Rupiah terhadap Mata Uang Riyal Ananda, Rahmadan Alam Ardan; Nazir, Alwis; Oktavia, Lola; Haerani, Elin; Insani, Fitri
Journal of Computer System and Informatics (JoSYC) Vol 6 No 3 (2025): May 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i3.7400

Abstract

Currency exchange rates play an important role in the economic stability of a country, especially in the context of international trade and global financial mobility. In Indonesia, fluctuations in the Rupiah exchange rate against the Saudi Arabian Riyal (SAR) have become a strategic issue, especially ahead of the Hajj season. This study aims to predict the exchange rate of Rupiah against Riyal in that period by using two forecasting approaches, namely Linear Regression and Trend Moment. The performance evaluation of both methods is conducted based on historical data using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) indicators. The results show that Linear Regression provides a better level of accuracy with an MAE of 330.36 and a MAPE of 17.32%, compared to Trend Moment which has an MAE of 412.41 and a MAPE of 18.88%. This finding shows that Linear Regression is more effective in capturing the pattern of exchange rate changes that tend to be linear. The prediction results also show an increasing trend in the exchange rate ahead of the Hajj month, which correlates with the increasing demand for foreign exchange. The implications of these results can be utilized by prospective pilgrims, business actors, and the government in formulating more appropriate and adaptive financial strategies