cover
Contact Name
Muhammad Khoiruddin Harahap
Contact Email
choir.harahap@yahoo.com
Phone
+6282251583783
Journal Mail Official
publikasi@itscience.org
Editorial Address
Medan
Location
Unknown,
Unknown
INDONESIA
Brilliance: Research of Artificial Intelligence
ISSN : -     EISSN : 28079035     DOI : https://doi.org/10.47709
Core Subject : Science, Education,
Brilliance: Research of Artificial Intelligence is The Scientific Journal. Brilliance is published twice in one year, namely in February, May and November. Brilliance aims to promote research in the field of Informatics Engineering which focuses on publishing quality papers about the latest information about Artificial Intelligence. Submitted papers will be reviewed by the Journal and Association technical committee. All articles submitted must be original reports, previously published research results, experimental or theoretical, and colleagues will review. Articles sent to the Brilliance may not be published elsewhere. The manuscript must follow the author guidelines provided by Brilliance and must be reviewed and edited. Brilliance is published by Information Technology and Science (ITScience), a Research Institute in Medan, North Sumatra, Indonesia.
Articles 544 Documents
Development of an Automatic Summarization System based on Large Language Models for Annual Report Analysis Rizki, Muhammad; Wibisono, Yudi; Nugroho, Eddy Prasetyo
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6772

Abstract

The increasing interest in stock market investment in Indonesia has highlighted a significant challenge for retail investors: the difficulty of analyzing lengthy and complex corporate annual reports. These documents, essential for fundamental analysis, are often hundreds of pages long and contain detailed narrative sections that require considerable time and effort to comprehend. This research addresses this issue by developing an automatic summarization system using a Large Language Model (LLM) to generate concise and insightful summaries of such reports. The primary objective was to develop and evaluate an LLM-based system specifically adapted for the structure and content of annual reports. The method involved creating a tailored dataset comprising 2,008 narrative text excerpts and their corresponding manual summaries sourced from the annual reports of companies listed on the Indonesia Stock Exchange (IDX). The open-source Llama-3.2-3B-Instruct model was then fine-tuned using the Parameter-Efficient Fine-Tuning (PEFT) technique, specifically Low-Rank Adaptation (LoRA). The research results demonstrated a significant improvement in the model's performance after fine-tuning. Quantitative evaluation using ROUGE metrics showed a relative increase of 18.63% in ROUGE-1, 44.45% in ROUGE-2, and 33.83% in ROUGE-L compared to the base model. Qualitative analysis confirmed that the fine-tuned model was capable of generating informative and relevant summaries aligned with the context of annual report analysis. In conclusion, this study demonstrates that fine-tuning LLMs with document-specific data is an effective approach for specialized tasks such as annual report summarization.
Non-Playable Characters Based On Large Language Models For Role Playing Games (RPG) Mulyana, Ade; Wibisono, Yudi; Anisyah, Ani
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6779

Abstract

Interactive dialogue is a central element in role-playing games (RPG), particularly those that emphasize storytelling and immersion. This study explores the development of a dynamic Non-Playable Character (NPC) system using a Large Language Model (LLM) to simulate responsive conversations in a fictional world. The objective of this research is to design an NPC dialogue system that can maintain contextual consistency with the game’s lore while adapting to player input dynamically. The method used is engineering-based development, involving prompt engineering and a Retrieval-Augmented Generation (RAG) approach to embed narrative context into the LLM prompts. The system is implemented in a 2D RPG titled Kage no Meiyaku: Shinobi no Michi, where players interact with multiple NPCs whose responses evolve based on both pre-defined lore and game progression. Evaluation is conducted using a Likert scale across four dialogue quality dimensions: coherence, emotional engagement, narrative relevance, and persona consistency. The results show that the system generates engaging and contextually accurate responses, with average scores ranging from 4.0 to 4.5. Some limitations are identified, such as occasional misspellings and generic responses in ambiguous inputs. However, the approach demonstrates strong potential for AI-assisted storytelling in games. This research contributes to expanding LLM applications in interactive fiction and opens future work toward feature-rich RPG elements such as transactional systems, branching narratives, and real-time battle interactions.
Benchmarking GPU Passthrough Performance on Docker for AI Cloud System Sani, Ahmad Faisal; Khoirunisa, Rifa; Riatma, Darmawan Lahru; Rachman, Yusuf Fadlila; Masbahah, Masbahah
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6794

Abstract

The use of artificial intelligence (AI), which depends only on CPU resources, tends to result in longer execution times or CPU time. Especially when handling large amounts or complex workloads. To overcome that issue, the use of a graphics processing unit (GPU) becomes a significant support. GPUs can significantly speed up AI inferences through their parallel architecture. One recent approach to integrating GPUs into an AI system is called GPU passthrough. Either natively (native environment), or through Docker environment. However, until recently, the efficiency and results between those methods have remained unexplored, particularly in local cloud environment.This study aimed to compare GPU performance between native and Docker environment using a 10.000 x 10.000 matrix multiplication workload with the TensorFlow frameworks. Execution time and GPU performance measured using the nvidia-smi tool. Data is recorded automatically in CSV format. The researcher used the NVIDIA CUDA environment to ensure full compatibility with GPU acceleration.The result demonstrated that GPU processing in native environment had faster average time, as in 1.52 seconds. In another case, GPU passthrough in docker environment demonstrated higher GPU utilization, as in 86.2% but had a longer execution time.These findings indicate that GPU overhead occurred in docker environment due to the containerization layer. On the contrary, the native environment resulted in shorter execution time, even though it did not maximize the GPU utilization. These results provide valuable basis data for technical decision-making in GPU-based AI deployment in a limited environment.
Optimizing Bit Depth for Longan Seedling Identification Using ANN and GLCM Andika, Rizki; Gasim, Gasim; Mair, Zaid Romegar
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6811

Abstract

Accurate identification of longan (Dimocarpus longan) seedling varieties is essential for agribusiness to select cultivars meeting market and environmental needs, but manual identification is error-prone due to similar leaf textures. This study optimizes grayscale image bit depth using Artificial Neural Networks (ANN) and Gray Level Co-occurrence Matrix (GLCM) to enhance longan seedling classification accuracy, addressing a gap in texture-based identification efficiency. Leaf images from five longan varieties (Itoh, Pingpong, Merah, Matalada, Diamond River) were captured with a USB digital microscope and converted to grayscale at bit depths of 4 (0–15), 5 (0–31), 6 (0–63), 7 (0–127), and 8 (0–255). Texture features (contrast, correlation, energy, homogeneity, entropy, standard deviation) were extracted using MATLAB. An ANN model, trained with the traingdx algorithm on 800 training and 200 test images, classified the varieties. The 6-bit and 4-bit depths yielded the highest accuracy (84.5%), followed by 7-bit (84.0%), 5-bit (83.5%), and 8-bit (82.0%), with Matalada achieving 90.0% accuracy. The 8-bit depth introduced texture noise, reducing performance. A 6-bit depth is optimal for longan leaf texture classification, though distinguishing similar varieties like Itoh and Pingpong remains challenging. Future research should incorporate color or morphological features to improve agricultural image processing.
Electronic Door App Development Using Machine Learning And Face ID Hidayat, Wisnu; Sri Lestari, Ninik; Gammanr, Dzulfikri Gammanr
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6816

Abstract

Face recognition technology (Face ID) has emerged as a highly secure solution by leveraging the distinctive attributes of each individual.  The application of Face ID in electronic door access systems significantly enhances security across residential, commercial, and public facilities.  This research's urgency can markedly enhance access security relative to traditional approaches and intelligent, secure, and efficient security solutions.  The objective of the research is to create a Face ID-based door security system, enhancing efficiency and reliability in access security.  The research employs the experimental technique alongside system development utilising the waterfall model.  The process involves analysing requirements to facilitate research, followed by system design, system testing, and concluding with system implementation.  Face ID employs artificial intelligence and machine learning technology to identify registered faces.  The precision of facial detection with a high-resolution camera and its connection with a smart lock may be managed via the application. A biometric authentication system utilizing facial recognition can serve as a substitute for traditional door lock mechanisms. The main components of the electronic door include biometric recognition based on Face ID. Its mechanism uses a solenoid lock to automatically control the door lock through electromagnetic action. The user interface is equipped with an LCD screen, which displays comprehensive information about the status of the electronic door.  The testing findings indicate that the input and output hardware of the system, specifically the camera and servo motor, function effectively.
Machine Learning System for Predicting Student Suitability for University Courses Wella, Rande; Sandra, Ahmadu A.; John, Anagu Emmanuel
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.5774

Abstract

The accuracy of student-course predictions determines both university learning success and career guidance process. When selecting students for courses using conventional methods universities fail to weigh all important factors which results in mismatching student-enrollment decisions. This research constructs a web-based machine learning framework to assess which university classes students would succeed in through fundamental attributes. The research conducts its analysis using background and academic data obtained from 2000 Taraba Secondary School students. The primary targets of this project involve understanding the course suitability determinants among students followed by building a machine learning model and creating a web-based interface and conducting a test on model effectiveness. Three machine learning algorithms—Random Forest, Decision Tree, and Support Vector Machine (SVM). A 95% accuracy emerged as the best outcome from using Random Forest whereas Support Vector Machine (SVM) achieved 93% accuracy and Decision Tree produced 92%. The predictive abilities for matching students to academic courses improve significantly through implementing machine learning algorithms. Students obtain automatic recommendations immediately after entering their student data on the platform. The new student guidance system shows students to proper courses ahead of time thus reducing curricular mismatches and boosting their academic outcomes. Future researchers must build adaptive learning models and real-time data updating functions to enhance accuracy and scalability levels in their work.
Information Security Risk Analysis Using ISO 31000:2018 and ISO 27001:2022 Ulya, Athiyatul; Karima, Annisa; Sukiman, T. Sukma Achriadi; Zulfia, Anni; Rahmawati, Rafika
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6564

Abstract

Information system risk audits are an important step in ensuring the security, effectiveness, and efficiency of the systems used by organizations. However, the fast advancement of information and communication technologies has made information?security threats more intricate, arising not only from internal sources like employee carelessness but also from external sources such as cyber?attacks, malware, and data?theft. This study aims to analyze information security risks at the Central Statistics Agency (BPS) of Lhokseumawe by referring to two international standards, namely ISO/IEC 27001:2022 and ISO 31000:2018. The research approach used is descriptive qualitative with a case study method. Data collection techniques were conducted through interviews, observations, and document studies. The results of the study indicate that there are still various security gaps, both technical and non-technical, such as weak system authentication, the absence of adequate security policies, and the lack of incident handling procedures. This study successfully compiled a risk register containing 30 types of risks along with their causes, impacts, likelihood levels, and relevant mitigation recommendations. Improvement recommendations include strengthening technical controls, updating information security policies, enhancing human resource capacity, and conducting regular internal audits. The results of this study are expected to serve as a reference for strengthening information security systems in a systematic and standardized manner within the BPS environment.
E-commerce as Marketing Media for Ruminant Feed and UMKM Craft Products Poto Village, Sumbawa Putraedi, Andri; Munandar, Imam; Hermanto, Koko; Yuliadi, Yuliadi; Julkarnain, M.; W, Yunari.; Akhir Putra, Juniardi
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6679

Abstract

This study aims to design, develop, and test an e-commerce prototype as a digital marketing medium for ruminant animal feed products and micro, small, and medium enterprise (MSME) crafts in Poto Village, Sumbawa Regency. Local business actors in this area generally face two main challenges: limited access to broader markets and the lack of optimal utilization of digital technology for product promotion and sales. The study employed a qualitative research method to address these issues with data collected through observation, interviews, questionnaires, and literature studies. The developed e-commerce platform was then tested to evaluate its effectiveness and usability. The results show that the platform contributes positively by expanding the marketing reach of local products, increasing their competitiveness, and simplifying the transaction process. In addition, users expressed favorable responses to the simple interface, the clarity of product display, and the integrated communication features that facilitate direct interaction between sellers and buyers. These findings highlight that implementing an e-commerce system tailored to local needs has significant potential to strengthen the rural economy, particularly through the digitalization of MSMEs and the livestock feed sector. Testing further confirms that all e-commerce features function properly, meet user requirements, and are very easy to operate, thus providing evidence of the system's practicality and sustainability for long-term community use.
Machine Learning Model for Human Resource Placement in Higher Education Syefudin, Syefudin; Kurniawan, Rifki Dwi
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6861

Abstract

This study presents the development and evaluation of a machine learning model designed to support human resource (HR) placement decisions in higher education institutions. Using combined personnel data from STMIK YMI Tegal and Politeknik Harber, we built a predictive model to estimate staff attendance at institutional progress reporting events, a critical indicator for performance evaluation and role suitability. The dataset comprised 137 records with six categorical predictors: Position, Homebase, Origin, Tegal_Status, Gender, and Institution. Categorical variables were encoded using label encoding, and a Random Forest classifier was trained using a stratified 75%/25% train-test split. The model achieved a held-out test accuracy of 97.14%, precision of 93.33%, recall of 100%, and F1-score of 96.55%, outperforming baseline models (Logistic Regression and Decision Tree). Five-fold cross validation confirmed robust generalization with an average accuracy of 91.22%. Feature importance analysis revealed Position as the most influential variable (76.88% importance), followed by Homebase and Origin. The results suggest that machine learning, particularly ensemble based methods, can provide reliable decision support tools for HR managers in academic settings, enabling data driven placement strategies. This research highlights the potential of predictive analytics for optimizing staff assignments and fostering institutional effectiveness. Future work should include larger datasets, additional features, and external validation to enhance model generalizability.
Smart Monitoring System For Hybrid Solar Power Plants Using Iot Technology Wirjawan, Agung; Safarudin, Andi; Ramdani, Rika; Sri Lestari, Ninik; Ramadi, Givy Devira; Sukirno, Sukirno
Brilliance: Research of Artificial Intelligence Vol. 5 No. 2 (2025): Brilliance: Research of Artificial Intelligence, Article Research November 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i2.6865

Abstract

Solar Power Plants (PLTS) have become one of the most widely used sources of renewable energy across various sectors, including both commercial industries and households. A hybrid solar power system combines solar energy with other sources, such as generators or the electrical grid.  The urgency of this research lies in the development of an Internet of Things (IoT)-based monitoring system for hybrid solar power plants, which can analyze data in real time, facilitate problem detection, optimize performance, and design preventive maintenance, as well as serve as an alternative source of electrical energy.The objective of this study is to develop a monitoring system for hybrid solar power plants using IoT. This technology is essential for enhancing the efficiency and sustainability of hybrid solar power generation. The method used in this research is the experimental method.Solar panels function as energy collectors that convert sunlight into electrical energy. The solar charge controller regulates the amount of electrical energy from the panels used to charge the battery. The voltage, current, and power generated by the plant are detected by two sensors: a DC voltage sensor and an ACS712 current sensor. The energy stored in the battery is used to power lamps. IoT offers an efficient solution for integrating components of a solar power plant, such as solar panels, inverters, batteries, and others. A hybrid solar power monitoring system can ensure maximum energy utilization and reduce long-term operational costs.