cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
muhammad siddik hasibuan
Contact Email
mhdsiddikhasibuan@gmail.com
Phone
-
Journal Mail Official
cosieaira@gmail.com
Editorial Address
Jl Pukat Banting IV NO 41 Medan Kecamatan Medan Tembung Kode Pos 20224
Location
Kota medan,
Sumatera utara
INDONESIA
Journal of Computer Science and Informatics Engineering
Published by Ali Institute of Research and Publication
ISSN : -     EISSN : 28278356     DOI : -
Core Subject : Science,
Computer Science & IT
Artificial Intelligence Machine Learning Natural Language Processing Computer Vision Text Speech Text Mining Data mining Cryptography Data visualization Expert System Deep Learning Fuzzy Logic IoT and smart environments Neural Networks Pattern Recognition Image Processing Optimization Digital Signal Processing Networking Technology Web intelligence
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 125 Documents
 9   10   11   12   13 
Detection of SSH Brute Force Attacks Based on System Logs in Debian 12 Serliana Barutu; Rasit Junaedi Silalahi; Lotar Mateus Sinaga
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1767

Abstract

The Secure Shell (SSH) service on Linux servers is frequently targeted by brute-force attacks. This study aims to outline the characteristics of authentication logs generated when the Debian 12 operating system faces such attacks. An experimental method was applied within a local network using a Debian 12 server as the target and Kali Linux as the attacker, with Hydra utilized as the tool to execute the attack. The experiments were conducted across three scenarios based on the volume of attack attempts: 10, 100, and 500 login attempts, while the Fail2Ban service was deactivated. Data analysis was performed quantitatively and descriptively, using Success Rate and Failed Login Rate metrics derived from records in journalctl and auth.log. The test results indicate that a higher attack volume corresponds to a larger amount of generated logs. In a scenario involving a success rate of 0.2%, only 1 password was accepted. The study also identified limitations in OpenSSH sessions (reaching the maximum authentication attempt limit), though Hydra was able to initiate new sessions automatically. Therefore, the system records detailed abnormalities without data loss, making it highly suitable as a primary foundation for server security audits
Prompt Engineering as a Digital Literacy Skill: A Conceptual Framework for Effective Human–AI Collaboration Maryam Hussaini; Mahmud Lawan; Ukasha Abubakar
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1775

Abstract

The rapid adoption of generative artificial intelligence (AI) tools has transformed the way individuals interact with digital technologies across educational, professional, and research contexts. Despite the growing reliance on AI systems, many users lack the competencies required to formulate effective prompts, evaluate AI-generated outputs, and engage in responsible human–AI collaboration. Existing digital literacy frameworks primarily emphasize information access, communication, and technology use, while providing limited attention to prompt engineering as a distinct digital competency. This research introduces the conceptual model, the PIERCE Framework, which places prompt engineering as a crucial digital literacy skill in the era of generative AI. The framework was conceptualized by analyzing the latest literature on digital literacy, AI literacy, human–AI interaction, and prompt engineering. The proposed framework is divided into six interrelated components: Prompt Intent Formulation, Intelligent Prompt Design, Engagement with AI, Response Evaluation, Continuous Refinement, and Ethical Application. Together, these components provide a structured pathway for effective human–AI collaboration. The framework contributes to the growing discourse on AI literacy by offering a comprehensive perspective on the competencies required for productive and responsible AI use. The findings highlight the importance of integrating prompt engineering into future digital literacy initiatives, educational programs, and professional development practices.
Comparative Analysis of Classification Algorithms for Predicting Student Examination Outcomes Based on Academic Datasets Imam Sugiharto; Sri Indri Wahyuni; Septian Tunijah Faradila; Yunita Yunita; Muhammad Ifan Rifani
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1780

Abstract

Current educational data utilization remains largely focused on final evaluations, resulting in delayed interventions for students at risk of academic failure. This study compares the performance of the Support Vector Machine (SVM), K-Nearest Neighbors (K-NN), and Random Forest algorithms in classifying student examination outcomes as the foundation of an Early Warning System. A total of 6,607 Student Exam Performance records were extracted and processed using Mode Imputation, Label Encoding, and StandardScaler. The experimental results indicate that the SVM model with an RBF kernel achieved the highest accuracy of 76.40%, outperforming Random Forest (73.22%) and K-NN (71.94%). This superior performance is primarily attributed to SVM's ability to construct precise decision boundaries in high-dimensional feature spaces. Therefore, the SVM model is recommended as the primary analytical engine for the early detection of potential academic failure in educational institutions
Evaluating User Experience and User Engagement of the AstraPay Mobile Wallet Using UEQ+ and UES Kelvin Isaiah Sitorus; Afen Prana Utama Sembiring; Yuni Marlina Saragih
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1786

Abstract

This study was conducted to see the extent of user experience (User Experience) and user engagement (User Engagement) on the AstraPay digital wallet application, considering that these two things are important keys that determine the success and sustainability of a digital application in the eyes of its users, where user experience is measured using the User Experience Questionnaire Plus (UEQ+) and user engagement is measured through the User Engagement Scale (UES), through a quantitative approach by distributing online questionnaires to 405 AstraPay users in Indonesia whose data is then analyzed using validity and reliability tests, calculating the average for each scale, Spearman correlation tests, to simple linear regression. The results, all UEQ+ scales show positive evaluations with Usefulness and Clarity recording the highest scores, while in the UES the Aesthetic Elements and Reward Factor scales performed very well, while Perceived Usability is the aspect with the lowest score. Regression analysis shows a determination coefficient value of R² = 0.251 with a regression coefficient B = 0.275 (p < 0.001), so that user experience has a positive and significant effect on user engagement, which means that the better the quality of experience perceived by users, the higher their level of engagement with the AstraPay application
Optimizing FTP Server Performance Using the Locality-Based Least Connection (LBLC) Algorithm in a Scheduling Algorithm Balancing System Ahmad Ridwan; Pramawahyudi Pramawahyudi; Enda Putri Atika; Budi Bayu Murti; Muzakki Ahmad
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1787

Abstract

A simultaneous increase in internet user traffic often causes servers to become overloaded, leading to disruptions, particularly on File Transfer Protocol (FTP) servers. A load-balancing system using Linux Virtual Servers is a solution for distributing traffic evenly. This study aims to analyze the performance of ten scheduling algorithms in a load-balancing system for File Transfer Protocol server applications with an Internet Protocol tunnel topology. The research method involves implementing a server cluster using the Debian operating system with one load-balancing server and two real servers. This topology allows the real servers to be located on geographically separate networks. Performance testing was conducted on ten different scheduling algorithms using five simultaneous clients, measuring response time and throughput using network analysis software. The test results showed that the Least Connections Based on Locality algorithm provided the most optimal performance compared to the other algorithms. The algorithm recorded the lowest average response time of 0.6066 seconds and the highest average throughput of 43 kilobits per second. These results are significantly better than those from tests without a load-balancing system, which yielded a response time of 2.7332 seconds. It can be concluded that the Least Connections Based on Locality algorithm is the most effective when applied to File Transfer Protocol servers with an Internet Protocol tunnel topology to improve network service quality.
Mapping Leading Commodities of Community Forest Plantations Based on Productivity Using the K-Means Clustering Algorithm Taufik Hidayat; Yuni Handayani; Muhammad Khozin; Tri Muji Waluyo; Dian Novitaningrum; Tresi Aprilia; Muchamad Achsin Samas
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1797

Abstract

Kendal Regency has significant potential in community forest plantations, which contribute to the regional economy. However, the mapping of leading commodities based on productivity has not been conducted optimally. This study aims to map leading community forest plantation commodities using the K-Means Clustering algorithm. The novelty of this study lies in the application of the K-Means Clustering algorithm by integrating land area and production volume as the basis for mapping leading commodities at the regency level. Secondary data from the Central Bureau of Statistics of Kendal Regency for the 2019–2023 period, covering seven community forest plantation commodities, were used. The research stages included data preprocessing using Min-Max normalization, clustering into three clusters using the K-Means algorithm, and cluster evaluation employing the Within-Cluster Sum of Squares (WCSS). The results show that the K-Means algorithm successfully grouped the commodities into three clusters based on their productivity characteristics. Sugarcane formed a distinct cluster as the leading commodity due to its highest productivity despite its relatively small cultivation area. These findings provide data-driven insights to support decision-making for the development of community forest plantations in Kendal Regency
Developing Statistical Data Analytics Competencies through AI-Assisted Interactive Dashboard Development Using Tableau in Vocational Higher Education Iswanti Iswanti
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1799

Abstract

The increasing demand for data-driven decision-making requires engineering graduates to possess strong statistical data analytics competencies. However, statistics learning in vocational higher education often provides limited opportunities for students to develop analytical competencies through authentic, AI-supported learning experiences. This study investigated the development of students' statistical data analytics competencies through AI-assisted interactive dashboard development using Tableau. A convergent mixed-methods design was employed involving undergraduate students enrolled in a Statistics course in the Informatics Engineering Department of Politeknik Negeri Semarang, Indonesia. Quantitative data were collected through competency-based assessment of students' interactive dashboard projects, while qualitative data were obtained through artifact analysis of three sequential authentic learning artifacts: AI-assisted authentic dataset development, interactive dashboard development, and AI-assisted analytical presentation. The findings indicate that students achieved proficient to advanced competency levels across technical, analytical, and professional competency dimensions. Qualitative findings further revealed that the three learning artifacts formed a progressive competency pathway, enabling students to move from knowing the data, to representing the data, and ultimately to reasoning and communicating with the data. These findings demonstrate that AI-assisted project-based learning provides an effective instructional approach for strengthening statistical data analytics competencies in engineering vocational higher education
Threshold Optimized Random Forest and XGBoost with SHAP for Low Birth Length Risk Prediction Rian andi saputra; Albert Yakobus Chandra
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1825

Abstract

Stunting is a chronic nutritional disorder that impairs children's cognitive development and longterm productivity, making early risk identification essential. However, previous studies have mainly focused on stunting classification or conventional machine learning models, with limited attention to integrating probability threshold optimization and explainable artificial intelligence (XAI) for low birth length risk prediction. This study proposes a predictive framework using Random Forest (RF) and XGBoost based on 249,626 valid records from the 2024 Indonesian Nutritional Status Survey (SSGI), where low birth length was defined as a birth length below 48 cm. The models were optimized through hyperparameter tuning, probability threshold optimization, Stratified 5-Fold Cross Validation, and SHAP analysis. Evaluation on 10,000 testing records showed that Random Forest accuracy increased from 76.76% to 78.58%, while XGBoost achieved the highest accuracy of 84.07% at a threshold of 0.79, although recall decreased to 27%. SHAP identified birth weight as the most influential predictor. These findings demonstrate that threshold optimization improves predictive accuracy, while SHAP enhances model interpretability to support early screening and clinical decision-making for low birth length risk
Quantum Natural Gradient vs. Adam Optimizer in Variational Quantum Classifiers: Crossover Analysis and Information Acquisition Efficiency Desi Amirullah; Lipantri Mashur Gultom
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1829

Abstract

Variational Quantum Circuits (VQCs) represent a central paradigm in near-term quantum machine learning, yet the comparative optimisation dynamics of quantum-aware and classical optimisers remain insufficiently characterised in realistic multi-class settings. We present a systematic empirical study comparing the Quantum Natural Gradient (QNG) optimizer against Adam within a VQC trained for ten-class digit recognition, employing eight qubits, three variational layers with RY-RZ-RX gate sequences, circular CX entanglement, and data re-uploading—yielding 72 trainable quantum parameters augmented by a classical linear readout head. A diagonal Quantum Fisher Information Matrix (QFIM) estimated via the parameter-shift fidelity metric underpins QNG, with a lazy update scheme (every five gradient steps) to equalise computational cost with Adam. Over 1,000 training epochs, Adam achieves 95% test accuracy while QNG achieves 92%, with QNG demonstrating markedly superior early convergence. Crossover analysis across four encoding-overlap bins confirms that QNG outperforms Adam exclusively in the low-overlap regime (ci < 0.54), consistent with the theoretical predictions of Kimura and Mitarai [10]. Information acquisition efficiency measurements reveal qualitatively opposite scaling behaviours: Adam’s Fisher-empirical gi scales as ci−2.19 , whereas QNG’s QFIM-fidelity gi is near scale-invariant (ci0.02). These findings provide actionable optimizer selection criteria for practical VQC deployments and offer the first empirical validation of the Kimura-Mitarai efficiency framework beyond the quantum phase estimation setting
Robustness of the YOLOv11n Model to Lighting Variations: A Case Study of Rice Leaf Disease Detection Rangga Putra Sopyan; Albert Yakobus Chandra
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1833

Abstract

Rice leaf diseases significantly reduce crop productivity, making accurate and reliable automatic detection essential for precision agriculture. Although recent YOLO-based approaches have achieved high detection accuracy under controlled conditions, the robustness of YOLOv11n against diverse lighting conditions in rice leaf disease detection remains insufficiently investigated, particularly under real-world field environments. This study evaluates the robustness of YOLOv11n using synthetic photometric transformations (S0-S11) and real field lighting conditions (L1-L4). Model performance was assessed using Precision, Recall, F1-Score, and mAP@0.5. Under baseline conditions, YOLOv11n achieved a Precision of 0.955, Recall of 0.936, F1-Score of 0.945, and mAP@0.5 of 0.949. The model remained highly robust to mild overexposure, mild underexposure, and saturation variations, while severe underexposure, extreme overexposure, and exposure scaling caused moderate to substantial degradation. Partial shadow produced the most severe performance decline, reducing mAP@0.5 by 90.20%. Furthermore, all real-world lighting scenarios exhibited performance drops exceeding 87%, revealing a pronounced domain gap between synthetic simulations and field conditions. These findings demonstrate that global photometric transformations alone are insufficient to represent complex real-world illumination, providing practical evidence for developing more robust lighting-aware object detection models for agricultural applications and emphasizing the importance of realistic robustness evaluation

Page 12 of 13 | Total Record : 125

 9   10   11   12   13 

Filter by Year

2022 2026


Reset
Filter By Issues
All Issue Vol 5 No 3 (2026): July Vol 5 No 2 (2026): April Vol 5 No 1 (2026): January Vol 4 No 4 (2025): October Vol 4 No 3 (2025): July Vol 4 No 2 (2025): April Vol 4 No 1 (2025): Januari Vol 3 No 4 (2024): October Vol 3 No 3 (2024): Juli Vol 3 No 2 (2024): April Vol 3 No 1 (2024): Januari Vol 2 No 4 (2023): Oktober Vol 2 No 3 (2023): Juli Vol 2 No 2 (2023): April Vol 2 No 1 (2023): Januari Vol 1 No 4 (2022): Oktober Vol 1 No 3 (2022): Juli Vol 1 No 2 (2022): April Vol 1 No 1 (2022): Januari More Issue