cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Chandro pardede
Contact Email
chandro.pardede@del.ac.id
Phone
+6282272752210
Journal Mail Official
jurnaldel@gmail.com
Editorial Address
-
Location
Unknown,
Unknown
INDONESIA
Journal of Intelligent Computing, Systems, and Applications (JICSA)
Published by Institut Teknologi Del
ISSN : 31231705     EISSN : 77312317     DOI : https://doi.org/10.54074/jicsa.v2i1
Core Subject :
Journal of Intelligent Computing, Systems, and Applications (JICSA) publishes high-quality research and review articles emphasizing practical applications, interdisciplinary insights, and technological innovation. The scope of the journal includes, but is not limited to, the following themes: Artificial Intelligence (AI), Computing, Electrical and Electronics Engineering, Data Science and Analytics
Arjuna Subject : -
Articles 15 Documents
 1   2 
The AI Race Is the New Gold Rush Samuel Tampubolon
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 01 (2025): September 2025
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i01.1

Abstract

The calls to frame artificial intelligence competition as a “new gold rush” have intensified as generative AI systems scale from niche laboratory research to billion-parameter open-source platforms. This review article revisits that metaphor through a data-science and data-analytics lens, to sharpen the earlier analysis with evidence from recent literature on educational disruption, regulatory pluralism, job market, and geopolitical spillovers. A mixed-methods research design combines the best of both bibliometric clustering and comparing study-cases from previous literature. Results confirm that the continued dominance of deep-learning stacks and vertical integration may reveal that an advantage is migrating toward high-trust data pipelines and adaptive governance. Techno-federal fragmentation in both the U.S and China fosters rapid experimentation but generates compliance friction. Meanwhile, the E.U.’s risk-averse-based AI Act attract “trust-seeking” healthcare pilots. In education, large language-model chatbots deliver personalized tutoring at scale and “Study and Learn Mode” yet still amplify academic-integrity concerns. It might be concluded that sustainable leadership will hinge less on raw computing power than on federated, privacy-preserving analytics that align with emergent social norms and regional regulations. Therefore, while also comparable to the 1970s space race era, today’s policy makers and data-science teams should co-design auditability, synthetic-data augmentation, and cross-border sandboxes to avoid a systematic “race to the bottom.”
Enhancing Drone-Based Multi-Object Tracking Through YOLOv10 and BoostTrack Integration Taufiqurrahman
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 01 (2025): September 2025
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i01.5

Abstract

This study investigates the integration of the YOLOv10 object detection framework with the BoostTrack tracking algorithm to enhance drone-based multi-object tracking (MOT). Leveraging the lightweight YOLOv10n model, the proposed system was evaluated against larger variants (YOLOv10s and YOLOv10m) across widely recognized MOT metrics, including IDF1, MOTA, precision, recall, and identity switches. Experimental results reveal that YOLOv10n consistently outperforms its larger counterparts, achieving near-perfect IDF1 and MOTA scores, high precision and recall, and zero identity switches. These findings demonstrate that smaller, computationally efficient models can deliver robust tracking accuracy and identity preservation in dynamic aerial environments, making them well-suited for real-time drone applications. The study underscores the potential of lightweight architectures in advancing scalable, efficient, and reliable drone-based surveillance systems
Decision Tree-Based Approach for Predicting Mental Health Symptoms in University Students: A Case Study at IT Del Rudy Chandra; Bella Wahmilyana Asril; Mhd Adi Setiawan Aritonang
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 01 (2025): September 2025
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i01.7

Abstract

Mental health plays a critical role in ensuring the well-being and productivity of university students. Institut Teknologi Del students encounter multiple stressors, including academic demands, dormitory life, and social interactions. This study proposes the development of a web-based predictive system to enable students to self-assess their mental health status using the Decision Tree algorithm. The dataset comprises 1,457 psychological screening records from IT Del students, categorized into two primary labels: Tend to Be Depressed and Tend to Be Anxious. The system architecture employs Python with Flask for the backend, Laravel for the frontend, and integration via a REST API. The development process follows the CRISP-DM methodology, while model performance is evaluated using a confusion matrix. For the Tend to Be Depressed category, the model achieved an accuracy of 0.931; precision: 0.931; recall: 0.931; F1-score: 0.931. For the Tend to Be Anxious category, performance reached accuracy 0.839; Precision 0.921; Recall 0.782; F1-score 0.846. With its straightforward yet effective model design, the proposed system offers a practical and accurate early detection tool for students’ mental health conditions. Furthermore, it has the potential to support campus counseling services in a digital, scalable, and sustainable manner
Image Super Resolution for Satellite Imagery fadjrianah; Bella Wahmilyana Asril; Hogan Eighfansyah Susilo
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 01 (2025): September 2025
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i01.9

Abstract

Super resolution (SR) techniques play a critical role in enhancing the quality of satellite imagery by reconstructing high resolution (HR) images from low resolution (LR) inputs. This study investigates the application of the Enhanced Super Resolution Generative Adversarial Network (ESRGAN) to improve the spatial resolution of satellite images. The ESRGAN framework integrates a Residual-in-Residual Dense Block (RRDB) generator, a Relativistic average GAN (RaGAN) discriminator, and an improved perceptual loss function to achieve sharper textures and more realistic structures. The 4× Satellite Image Super Resolution dataset, consisting of paired LR (2 m/pixel) and HR (0.5 m/pixel) images, was employed. Quantitative evaluation was conducted using the Peak Signal-to-Noise Ratio (PSNR), with Lanczos5 interpolation used as a baseline for comparison. Experimental results on ten test images demonstrate that ESRGAN consistently outperforms Lanczos5. Qualitative assessments further confirm that ESRGAN effectively restores high frequency details, enhances fine textures, and preserves object boundaries more accurately than the interpolation method. These findings highlight ESRGAN’s potential as an advanced solution for improving visual quality and detail preservation in high resolution satellite imagery applications.
Development of a Web-Based Alumni Information System for Career Tracking and Alumni Networking Using Graph Network Algorithm: Case Study at Fakfak State Polytechnic Kezia Basary; Apryan; Rahmayani; Ardhyansyah Mualo
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 01 (2025): September 2025
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i01.12

Abstract

The web-based Alumni Information System is a digital solution designed to facilitate alumni career transitions and foster interpersonal relationships. This system is developed using the Object-Oriented Programming (OOP) method, which makes the code structure more modular, organized, and easy to use in long-term development. The main features of this system are alumni registration, filling and updating alumni work history, searching for alumni based on certain criteria, and a communication forum. Each component of the system can be tested independently and is highly integrated with OOP. The system performance is guaranteed to be stable and functions according to user needs. This system not only helps institutions manage their undergraduate careers; but also fosters professional collaboration among alumni. This system is expected to be effective in creating an active and contributing alumni community.
Comparative Analysis of Cross-Site Scripting (XSS) Vulnerabilities in Monolith and Microservices Architectures Evi Damanik
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 02 (2026): March 2026
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i02.14

Abstract

This study presents a comparative analysis of Cross-Site Scripting (XSS) vulnerabilities between monolithic and microservices web application architectures. Both applications were developed using Python Flask inside an isolated Docker environment and implemented with identical functionality. Security testing was conducted using automated scanning with OWASP ZAP on both architectures. The results show that Reflected XSS and Server-Side Template Injection (SSTI) vulnerabilities were detected in both architectures, with identical payloads successfully executed. These findings indicate that architectural separation alone does not automatically eliminate source-code-level vulnerabilities such as XSS. Security implementations must be consistently applied across all service layers to prevent data contamination and maintain trust boundaries between services.
Comparative Analysis of Collaborative Filtering Models on Highly Sparse Book Datasets Mickael Sitompul; Jeremy Nainggolan; Reinaldy Hutapea; Tennov Pakpahan; Ioka Purba
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 02 (2026): March 2026
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i02.19

Abstract

The rapid growth of digital information necessitates robust recommender systems to filter relevant content effectively. This study addresses the challenge of generating personalized book recommendations using a highly sparse dataset (>99% sparsity) derived from the Recommender System 2023 Challenge. The primary objective is to predict user interest based on implicit feedback and optimize the ranking quality of Top-N recommendations. We conduct a comparative analysis between a memory-based baseline, Item-Based Collaborative Filtering (IBCF), and a model-based approach, Alternating Least Squares (ALS). The methodology incorporates k-core decomposition for data preprocessing to mitigate sparsity issues, followed by rigorous hyperparameter optimization via grid search to tune latent factors, regularization parameters, and confidence weights. Experimental results demonstrate that the baseline IBCF model struggles significantly with data sparsity, yielding a Mean Average Precision at 10 (MAP@10) of only 0.0018. In contrast, the hyperparametertuned ALS model achieves a MAP@10 of 0.0348, representing a 19.3- fold improvement over the baseline and a 3.3-fold improvementover the default ALS configuration. These findings confirm thatmatrix factorization techniques, when systematically optimized,significantly outperform memory-based methods in handling sparseimplicit feedback, providing a scalable and accurate solution forranking tasks.
The Comparative Analysis of Item-Based Collaborative Filtering and Alternating Least Squares on Implicit Feedback Recommender Systems Dolok Butarbutar; Brain Sitorus; Krisnia Siahaan; Bryan Simamora; Nikita Simanjuntak
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 02 (2026): March 2026
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i02.21

Abstract

This study addresses the challenge of predicting user preferences using implicit feedback data. We compared Popularity-Based and Item-Based Collaborative Filtering (IBCF) baselines against the Alternating Least Squares (ALS) method, evaluating performance via Mean Average Precision at 10 (MAP@10). Results show that ALS- Tuned achieved the highest score (MAP@10 = 0.012788), showing best performance compared to the baseline model. This indicates that hyperparameter tuning affects the most in model performance.
Top-N Recommendation (Implicit Feedback) Bethania Hasibuan; Rahel Uli Rotua Simanjuntak; Christoffel Teofani Napitupulu; Chika Situmorang; Difya Laurensya Ambarita
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 1 No. 02 (2026): March 2026
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v1i02.22

Abstract

The rapid growth of digital platforms has led to a massive increase in user-item interaction data, most of which is recorded as implicitfeedback such as purchase history and reading duration. Unlike explicit ratings, implicit feedback presents fundamental challengesincluding extreme sparsity and ambiguity, as non-interaction does not necessarily indicate disinterest. This study develops a Top-Nbook recommendation system by implementing Item-Based Collaborative Filtering (IBCF) as a baseline and Alternating LeastSquares (ALS) as the primary model, addressing the challenge of a highly sparse dataset (13,876 users × 123,069 items). Themethodology involves transforming raw interaction data into Compressed Sparse Row (CSR) format, performing systematichyperparameter tuning on the ALS model, and implementing rigorous data preprocessing techniques to optimize performance.The experimental results demonstrate a progressive improvement trajectory: the baseline IBCF model achieved a Mean AveragePrecision at 10 (MAP@10) score of 0.0186, the initial ALS model improved to 0.0243, hyperparameter tuning further increasedperformance to 0.0257, while the integration of systematic data preprocessing (K-Core filtering) with advanced hyperparametertuning ultimately yielded a score of 0.0622 representing a 234% improvement over the baseline. Qualitative analysis reveals thatwhile IBCF produces monotonous and homogeneous recommendations limited to similar publishers and series, ALSprovides substantially more diverse and exploratory results by capturing latent behavioral patterns across multiple genres andthematic categories. This study concludes that the optimized ALS model, combining matrix factorization with rigorous dataengineering, is significantly more effective than in handling sparse implicit feedback, delivering superior rankingaccuracy, enhanced recommendation diversity, and improved user discovery experience.
Performance Comparison of Recommendation Models: IBCF and ALS Based on Implicit Feedback Samuel Sitio; Wilson Sihombing; Winfrey Nainggolan; Bronson Siallagan; Reinhard Batubara; Wiwin Sry Adinda Banjarnahor
Journal of Intelligent Computing, Systems, and Applications (JICSA) Vol. 2 No. 1 (2026): September 2026
Publisher : Penerbit Institut Teknologi Del

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54074/jicsa.v2i1.23

Abstract

The rapid growth of digital platforms has led to an overwhelming amount of information, making it difficult for users to discover relevant books among the vast selection available. Recommendation systems, particularly collaborative filtering techniques, help address this issue by personalizing recommendations based on user interactions. This study compares the performance of three recommendation models: Item-Based Collaborative Filtering (IBCF) as a baseline, Alternating Least Squares (ALS), and ALS with hyperparameter tuning (ALS-Tuning) in the context of implicit feedback data. Using two evaluation metrics, Mean Average Precision at 10 (MAP@10) and Root Mean Squared Error (RMSE), the results demonstrate that ALS-Tuning outperforms both IBCF and the standard ALS model in both metrics. The ALS-Tuning model shows the highest improvement in ranking relevance (MAP@10) and prediction accuracy (RMSE), with hyperparameter tuning playing a significant role in enhancing the model's ability to uncover latent user preferences. While IBCF struggles with sparse data, ALS, especially with tuning, proves to be more effective in capturing latent patterns in implicit feedback data. This research provides empirical evidence supporting the use of ALS-Tuning as a robust approach for recommendation systems with sparse and implicit data, offering valuable insights for future development in personalized recommendation systems.

Page 1 of 2 | Total Record : 15

 1   2 

Filter by Year

2025 2026


Reset
Filter By Issues
All Issue Vol. 1 No. 02 (2026): March 2026 Vol. 2 No. 1 (2026): September 2026 Vol. 1 No. 01 (2025): September 2025