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 Articles sent to the Brilliance may not be published elsewhere.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
Imputing Data and Predicting Waste with Machine Learning in East Java Khoirunisa, Rifa; Sani, Ahmad Faisal; Riatma, Darmawan Lahru; Masbahah, Masbahah; Rachman, Yusuf Fadlila
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6461

Abstract

Indonesia's waste problem continues to be a pressing environmental issue, along with the increasing population and urbanization activities. The increase in population and changes in consumption patterns have led to a significant spike in waste generation in Indonesia. Machine learning-based approaches become highly relevant in supporting accurate predictive systems to estimate waste generation, so that it can be used as a basis for policy making and planning for more effective and sustainable waste management. However,the issue of missing data is a common challenge in environmental data processing, including in the recording of waste generation. Incomplete waste generation data can hinder accurate analysis and prediction, which are essential for effective environmental management planning. This study aims to analyze the effectiveness of various data imputation methods and to develop a predictive model for waste generation in East Java Province using a machine learning approach. The imputation techniques tested include Mean Imputation, K-Nearest Neighbor (KNN), and Interpolation, while the predictive models used include Random Forest, Gradient Boosting, and KNN Regression. The dataset was obtained from the official SIPSN (National Waste Management Information System) website. Model performance was evaluated using metrics such as Root Mean Square Error (RMSE). The results indicate that the combination of KNN Imputer with the Gradient Boosting prediction model is effective in addressing missing data and predicting waste generation in East Java Province, achieving an RMSE value of 0.147. These findings are expected to support more accurate decision-making in waste management planning for the province.
Performance Comparison of SelectKBest and Permutation Importance in Feature Selection for Diabetes Prediction Cahyani, Nita; Irsyada, Rahmat
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6507

Abstract

This study evaluates the effectiveness of two feature selection methods, namely the statistics-based SelectKBest and the model-based Permutation Importance, in improving the performance of classification algorithms for diabetes prediction. A dataset consisting of 17 clinical and demographic features was used to train 11 machine learning algorithms with two subsets of selected features. Performance evaluation used accuracy, precision, recall, F1-Score, ROC AUC, and training time. Based on the results, the SelectKBest method was able to improve the performance of Random Forest with an accuracy of 82.7%, a precision of 0.8, a recall of 0.5, and an F1-Score of 0.615. Meanwhile, the Permutation Importance method showed more consistent performance, with six models including Random Forest, K-Nearest Neighbors, and Quadratic Discriminant Analysis (QDA) achieving an accuracy of up to 86.2%. QDA stood out with the highest ROC AUC of 0.887, indicating better class detection capabilities. These findings underscore the superiority of Permutation Importance in selecting relevant and varied features, including demographic factors, thereby improving overall prediction accuracy. In practice, Random Forest with SelectKBest is recommended for applications requiring fast and interpretable models, while QDA and Gradient Boosting with Permutation Importance are recommended for those requiring high accuracy and sensitivity. This study strengthens the foundation for developing more accurate and applicable diabetes prediction models across various contexts.
Designing an Information System for Student Admissions at SMAN 1 Pademawu Using the Waterfall Method Putra, Fauzan Prasetyo Eka; Mu'minin, Fadilatul; Nuraini, Alief; Barokah, Selviana Nur Rizqi; Khairurrozi, Khairurrozi
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6508

Abstract

 New Learner Admission (PPDB) is one of the processes of accepting students at every school level, starting from PAUD, kindergarten, elementary school, junior high school, to high school / vocational school. This term is used by various school bodies when they want to accept new students. Often these institutions experience problems in managing the administration of new participants, one of which is SMAN 1 Pademawu. The obstacles that occur include the buildup of queues of participants who register at the institution and constraints in inputting new participant data due to the limited number of PPDB organizing committee members. Therefore, this research will create a PPDB information system to overcome the obstacles experienced by the institution. The information system created is in the form of flow chart design, ERD (Entity Relationship Diagram), UML (Unified Modeling Language) and DFD (Data Flow Diagram). Using Javascript programming language, React Js framework and bootstrap. The result of this research is the construction of PPDB information system at SMAN 1 Pademawu by using waterfall method. With this system, it is expected to facilitate the PPDB committee of SMAN 1 Pademawu in managing or inputting data and reducing the buildup of registration queues effectively and efficiently.  
Creating a Tourist Destination Route Database as a Tourism Information Service in West Sumbawa Regency Hamdani, Fahri; Yuliadi, Yuliadi; Nuryadi, Halid; Atmawan Oktavia, Siska; Rosika, Herliana; Dzil Ikram, Fadhli
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6558

Abstract

West Sumbawa Regency is one of the regencies on Sumbawa Island, West Nusa Tenggara Province. Tourism is a popular activity; most people use it to fill their free time amid their busy lives. Tourism is vital in driving national economic growth because it is one of the development sectors that can spur economic growth in a region. The impact of current technological advances has resulted in all types of information being presented online. The West Sumbawa Regency Government has problems and difficulties conveying information about existing tourism because there are no information media that the public can easily access. Currently, West Sumbawa Regency does not have a tourist route search application that tourists use as an alternative to finding travel routes by tourists. Based on these conditions, media is needed to find tourist routes and a medium for promoting tourism and its products to tourists. It can make it easier for tourists to find the shortest route to a more efficient tourist spot. In addition, local government and tourism industry entrepreneurs are used as media to promote tourist attractions and their products and tourist needs. A tourist route database was developed using the waterfall method, PHP programming, and the MySQL database as a database management system (DBMS). The results are a tourist destination route database and a website-based tourism information service expected to make it easier for the public to obtain information about tourist routes in West Sumbawa Regency.
Deep Learning for Classifying Tenera and Dura Oil Palm Using ResNet-50 Putri, Anisah Dhiyaa Azzahra; Tinaliah, Tinaliah
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6562

Abstract

The plantation sector significantly contributes to Indonesia’s economy, with oil palm being a leading commodity in both domestic and international markets. Accurate identification of oil palm fruit varieties, particularly Dura and Tenera, is crucial for maximizing productivity and profitability. However, conventional identification methods are still manual, time-consuming, and prone to human error. This study proposes an automated classification approach using the ResNet-50 deep learning architecture to classify oil palm fruit varieties, and compares the performance of two optimizers: ADAM and SGD. A dataset of 2,000 images representing Dura and Tenera varieties was collected and augmented to enhance training diversity. The images were preprocessed and resized to 224×224 pixels before being input into the model. Experiments were conducted with variations in learning rate, batch size, and dense layers. Model performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrix. The results show that the ResNet-50 model trained with the ADAM optimizer achieved the best performance, with 85% testing accuracy and an F1-score of 0.85, outperforming the SGD optimizer. These findings demonstrate that combining ResNet-50 with ADAM results in better generalization and training efficiency. This research supports the development of a more accurate and efficient classification system for oil palm management, with potential for broader application in agricultural automation.
Classification of Mango Varieties from Leaf Images Using ResNet-50 CNN Architecture Komah, Neilsen Nicholas; Hermanto, Dedy
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6571

Abstract

Early identification of mango (Mangifera indica L.) varieties is crucial for optimizing cultivation, as each variety possesses distinct characteristics and regional adaptability. However, the morphological similarities in leaf shape and texture especially among Harumanis, Erwin, Cokanan, Gedong Gincu, and Mahatir pose challenges for novice farmers and hobbyists. This study proposes a classification system using the Convolutional Neural Network (CNN) method with the ResNet-50 architecture to classify mango leaf varieties based on image data. A total of 5,000 images were collected and augmented from 1,250 original samples using a high-resolution camera under controlled indoor conditions. The dataset was split into training (80%), validation (10%), and testing (10%). Sixteen experimental configurations were evaluated using combinations of image resolutions (160×160 and 320×320 pixels), learning rates (0.01, 0.001), batch sizes (16, 32), and training epochs (50, 100). The best results were achieved using a 320×320 image size, learning rate of 0.001, batch size of 32, and 100 epochs, yielding a validation accuracy of 89.9%, precision of 89.87%, recall of 89.9%, and F1-score of 89.83%. These results confirm that high-resolution images and fine-tuned hyperparameters significantly enhance classification performance. The findings demonstrate the effectiveness of the ResNet-50 model for fine-grained classification in agriculture and support its future deployment in real-world environments for cultivar identification, quality control, and intelligent crop management.
Design of an Automatic Help Desk Response Module Using Natural Language Processing Fridayanto, Tri Mur; Wibowo, Ari Purno Wahyu
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6575

Abstract

Manual help desk systems in enterprise environments often suffer from delayed response times and repetitive queries, reducing service efficiency. This research aims to design an automated help desk response module by applying Natural Language Processing (NLP) techniques, specifically within the asset management context of an ERP system. The module uses Term Frequency-Inverse Document Frequency (TF-IDF) and cosine similarity to classify incoming queries and retrieve relevant answers from a predefined knowledge base. Python, Django, PostgreSQL, Scikit-learn, and NLTK were used to implement the module. Testing was conducted using 50 sample queries, resulting in an accuracy of 90% based on confusion matrix evaluation. The system successfully retrieves appropriate responses for most frequent user issues. This design is expected to support organizations in streamlining their help desk operations and improving response time and consistency. Future developments may involve semantic matching and machine learning-based improvements to enhance understanding of unstructured queries.
Sentiment Analysis of Mental Health Using Support Vector Machine (SVM) with FastAPI Implementation Maulyanda, Maulyanda; Sri Azizah Nazhifah
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6580

Abstract

Mental health is a vital aspect that contributes significantly to an individual’s productivity, daily activity, and overall quality of life. With the increasing prevalence of mental health issues, early detection and analysis are essential. This study aims to identify mental health conditions using a machine learning approach, specifically the Support Vector Machine (SVM) algorithm. The dataset used consists of 53,043 text-based statements that are classified into seven distinct categories of mental conditions: normal, depression, suicide, anxiety, bipolar, stress, and personality disorders. The preprocessing steps applied to the dataset include text cleaning, tokenization, stopword removal, and lemmatization to standardize the textual input. Following this, 80% of the data is allocated for training the model, while the remaining 20% is reserved for testing purposes. The SVM algorithm is utilized to build a predictive model capable of classifying mental conditions based on text input. Furthermore, this model is deployed through an application interface using the FastAPI framework, enabling integration with digital platforms. The results indicate that the model achieves an accuracy of 79%, a recall of 77%, and an F1-score of 73%. These findings suggest that SVM is a capable and efficient method for analyzing and detecting various mental health conditions. This approach supports early intervention and offers practical implications for digital mental health screening tools.
Challenges in the Implementation of Interactive Multimedia Using the Discovery Learning Model to Enhance Logical Thinking Adeliani, Cindy Sri Meidina; Wahyudin, Wahyudin; Megasari, Rani
Brilliance: Research of Artificial Intelligence Vol. 5 No. 1 (2025): Brilliance: Research of Artificial Intelligence, Article Research May 2025
Publisher : Yayasan Cita Cendekiawan Al Khwarizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/brilliance.v5i1.6583

Abstract

In facing the challenges of the digital era, education in Indonesia must innovate through the integration of technology to produce a superior workforce. Traditional teacher-centered teaching methods have become an obstacle to developing students' logical thinking skills in problem-solving and conclusion-drawing. This research aims to develop and test the effectiveness of website-based interactive multimedia using the discovery learning model on the Entity Relationship Diagram (ERD) topic to enhance students' logical thinking skills. The media development employed the ADDIE model (Analysis, Design, Development, Implementation, Evaluation), with the interactive multimedia flow designed according to the stages of discovery learning. Effectiveness was measured using a one-group pretest-posttest design with vocational high school students in West Bandung, while the feasibility of the material and media was validated by experts. The improvement in logical thinking skills was measured using the N-gain test. The results showed that the developed interactive multimedia was declared highly feasible by material and media experts. The effectiveness test indicated an improvement in students' logical thinking skills, with an N-gain score in the medium category. However, a detailed breakdown of the improvement per aspect revealed that argumentation skills were moderate, while thought coherence and conclusion-drawing skills remained low. These difficulties were attributed to a lack of student initiative and motivation in discussion and summarizing the material. It is concluded that this interactive multimedia is feasible and effective as an alternative learning medium for training students' logical thinking skills, although further optimization is needed to improve thought coherence and conclusion-drawing abilities.
Implementation of AI in Animated Health Education Film to Overcome Hospital Anxiety Setiawan, Rudi; Nugroho, Nurhasan; Pratama, Gerald Untirtha
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.v5i2.4700

Abstract

Children's fear of going to hospitals is a common emotional barrier that can interfere with medical treatment and early health education. This research addresses that issue by developing an animated educational film using artificial intelligence (AI) to help children aged 6–9 overcome their anxiety about medical environments. The purpose of this study is to produce a multimedia-based health education tool that is emotionally engaging, educationally effective, and technically efficient. The film development followed the Multimedia Development Life Cycle (MDLC), consisting of six stages: concept, design, material collecting, assembly, testing, and distribution. Several AI-based tools were integrated, including ChatGPT for scriptwriting and visual prompting, Kling AI for animation and lip-syncing, ElevenLabs for AI-generated voice-over, and CapCut for final editing. The result is a 7-minute 2D animation titled "Mengatasi Ketakutan Berobat ke Rumah Sakit", which introduces child-friendly characters and a relatable storyline to reduce fear and improve comprehension of hospitals. User testing was conducted on five children aged 8–12, showing a noticeable decrease in hospital-related fear and increased verbal engagement with the topic. Most participants found the animation enjoyable and easy to follow. The findings suggest that AI-supported animation can be an effective medium for health education in children. While the outcome met the expected goals, future research is recommended to include larger participant groups and validated psychological assessment tools for more robust evaluation.