cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
+6282370070808
Journal Mail Official
jurnal.bulletincsr@gmail.com
Editorial Address
Jalan sisingamangaraja No 338 Medan, Indonesia
Location
Kota medan,
Sumatera utara
INDONESIA
Bulletin of Computer Science Research
Published by Forum Kerjasama Pendidikan TInggi
ISSN : -     EISSN : 27743659     DOI : -
Core Subject : Science,
Computer Science & IT
Bulletin of Computer Science Research covers the whole spectrum of Computer Science, which includes, but is not limited to : • Artificial Immune Systems, Ant Colonies, and Swarm Intelligence • Bayesian Networks and Probabilistic Reasoning • Biologically Inspired Intelligence • Brain-Computer Interfacing • Business Intelligence • Chaos theory and intelligent control systems • Clustering and Data Analysis • Complex Systems and Applications • Computational Intelligence and Soft Computing • Distributed Intelligent Systems • Database Management and Information Retrieval • Evolutionary computation and DNA/cellular/molecular computing • Expert Systems • Fault detection, Fault analysis, and Diagnostics • Fusion of Neural Networks and Fuzzy Systems • 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 • Knowledge discovery and ontology engineering • Machine Learning, Reinforcement Learning • Networked Control Systems • Neural Networks and Applications • Natural Language Processing • Optimization and Decision Making • Pattern Classification, Recognition, speech recognition, and synthesis • Robotic Intelligence • Rough sets and granular computing • Robustness Analysis • Self-Organizing Systems • Social Intelligence • Soft computing in P2P, Grid, Cloud and Internet Computing Technologies • Support Vector Machines • Ubiquitous, grid and high-performance computing • Virtual Reality in Engineering Applications • Web and mobile Intelligence, and Big Data • Cryptography • Model and Simulation • Image Processing
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 19 Documents
 1   2 
Search results for , issue "Vol. 6 No. 2 (2026): February 2026" : 19 Documents clear
Lightweight Convolutional Neural Network Based on Modified LeNet for Retinal Pathology Classification in High-Resolution Fundus Imaging Mu’awanah, Cahyatul; Hakim, Lukman
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.796

Abstract

Eye disease are visual impairments that can lead to blindness if not detected early. Fundus imaging is one of the most effective methods for identifying abnormalities in the eye. With the advancement of deep neural network technologies, particularly Convolutional Neural Network (CNN), the classification of fundus image can now be performed efficiently. LeNet is a well-known CNN architecture commonly used in image classification tasks, however it has limitation when processing images with complex visual features with high resolution, such as fundus images. This study proposes a modification to the LeNet architecture to enhance it’s a ability to extract important features from images with high resolution. The modification involves adding convolutional layers and adjusting image resolution to optimize the models performance in detecting eye disease in fundus images. The dataset used consists of 4,217 fundus images, classified into four categories: normal, cataract, glaucoma, and diabetic retinopathy. Experimental result show that the original LeNet-5 achieved an accuracy 0f 76%, while the modified LeNet architecture improved the accuracy to 86%. The main contibution of this research lies in the development of a modified and lighweight LeNet architecture, which is capable of handling high-resolution fundus images while maintainig computational efficiency and producing better classification performance compared to the original LeNet.
Model Deteksi Berita Hoaks Bahasa Indonesia Menggunakan Multinomial Naïve Bayes dan AdaBoost Classifier Hafiizh, Haniifaa; Juanita, Safitri
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.927

Abstract

The rapid growth of the internet has led to the massive and uncontrolled dissemination of information across various digital platforms, allowing hoax news to reach a wide audience and influence public opinion in a short period of time. This condition highlights the need for a reliable automated detection system. However, existing methods still face limitations in terms of accuracy, result stability, and reliance on manual verification processes. Therefore, this study aims to compare and analyze the performance of two classification algorithms in detecting Indonesian-language hoax news accurately and effectively. This study follows the CRISP-DM framework, beginning with the collection of hoax and non-hoax news articles from turnbackhoax.id and detik.com, resulting in 2,281 samples. The data understanding stage involves analyzing dataset characteristics and evaluating data quality. During data preparation, text elements that explicitly indicate hoax labels are removed, followed by feature extraction using Term Frequency–Inverse Document Frequency (TF-IDF). The dataset is then trained and tested using data split ratios of 70:30, 80:20, and 90:10 by applying Multinomial Naïve Bayes and AdaBoost Classifier algorithms. Model performance is evaluated using a confusion matrix. The results show that AdaBoost achieves superior performance, with an accuracy of 0.9879 (98.79%), outperforming Multinomial Naïve Bayes, which attains an accuracy of 0.9712 (97.12%). The performance of AdaBoost is also consistent across different evaluation scenarios, indicating that it is more suitable as an automated hoax news detection model for the dataset used in this study.
Analisis Tingkat Sentimen Opini Publik Terhadap Kebijakan TV Digital di Platform X Menggunakan Multinomial Naïve Bayes Sulaeman, Asep Arwan; Naya, Candra; Danny, Muhtajuddin; Effendi, M. Makmun
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.951

Abstract

The migration from analog to digital television broadcasting is part of the transformation of the broadcasting system aimed at improving broadcast quality and spectrum efficiency. However, the implementation of the digital television policy has generated diverse public responses, ranging from support to criticism. This study aims to analyze public opinion on the digital television policy in Indonesia using social media data from platform X. A quantitative approach was employed using text mining and supervised machine learning techniques. Data were collected through a crawling process using the keyword “tv digital”, resulting in 1,855 tweets. After data selection and cleaning, 789 tweets were obtained as the final dataset. The analysis stages included text preprocessing, feature extraction using Term Frequency–Inverse Document Frequency (TF–IDF), and sentiment classification using the Multinomial Naïve Bayes algorithm. The results indicate that positive sentiment dominates public opinion, with 478 tweets (60.58%), while negative sentiment accounts for 311 tweets (39.42%). Model performance evaluation shows an accuracy of 79.21%, precision of 82.45%, and recall of 85.06%, indicating that the model performs well and consistently in classifying sentiment. These findings demonstrate that social media–based sentiment analysis can serve as an empirical approach to understanding public perceptions of digital television policy.
Analisis Pengaruh Preprocessing Data dan Hyperparameter Tuning pada Backpropagation Neural Network dalam Klasifikasi Stroke Gunawan, Asrul; Hermawan, Arief; Avianto, Donny
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.956

Abstract

Data imbalance and scale differences between features are often the main factors that reduce the performance of neural network-based classification models. This study aims to analyze the effect of data preprocessing and hyperparameter tuning on the performance of Backpropagation Neural Network (BPNN) in stroke classification. This study used a stroke dataset from the Kaggle platform consisting of 5,110 patient data with 10 clinical features. The evaluation was conducted using five schemes and consisted of several data balancing techniques. These techniques include no balancing, SMOTE, and ADASYN. In addition, the evaluation also involved data normalization including no normalization, MinMaxScaler, and Z-Score. The BPNN model used has an architecture of 19 input neurons, 29 neurons in the hidden layer, and 1 output neuron. Hyperparameter tuning was performed by finding the best learning rate and number of epochs. The evaluation results showed that the model in scheme one has limitations. This limitation is most visible in identifying stroke classes. The application of SMOTE and MinMaxScaler in scheme two proved that the results were better and its performance increased significantly. On the other hand, the combination of ADASYN and Z-Score in scheme three showed more stable performance and was able to detect stroke cases more accurately. The hyperparameter tuning process in schemes four and five also proved to improve performance. The best results were obtained in scheme five, with an accuracy of 96.47%, a precision of 97.34%, a recall of 95.62%, and an F1-score of 96.47%. These findings indicate that the combination of adaptive balancing techniques, distribution-based normalization, and optimal parameter tuning is very effective in improving the accuracy and stability of BPNN for stroke classification.
Rute Terpendek Pengiriman Katering Makanan Menggunakan Geographic Information System dengan Metode Dijkstra Adira, Muhammad Faris; Triase, Triase
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.957

Abstract

The problem of determining catering delivery routes at UMKM Vfoodia in Medan City is still carried out manually and relies heavily on couriers’ experience, which may lead to inefficiencies, especially in cases of courier replacement and limited delivery time windows. This condition results in delivery delays and difficulties for couriers in understanding customer locations and delivery sequences. This study aims to develop a catering delivery route determination system based on Geographic Information System (GIS) using the Dijkstra algorithm. The system is developed as a web-based application accessible via Android devices to support both administrative and courier activities. GIS is utilized to visualize customer locations and road networks on a digital map, while the Dijkstra algorithm is applied to compute the shortest route between two points. In daily delivery operations involving multiple destinations, the Dijkstra algorithm is executed repeatedly, where the destination point is updated each time a customer delivery is completed. The system is integrated with the OpenRouteService API to obtain distance and travel time estimations and is equipped with a caching mechanism to reduce repetitive API calls. The contribution of this research lies in the application of the standard Dijkstra algorithm in a repetitive manner within a GIS-based system to support structured multi-destination catering delivery at the UMKM scale. Experimental results show that the system is able to generate the shortest delivery route with a minimum distance of 6.0 km in the test scenario and helps make the delivery process more organized and easier for couriers to understand. Therefore, the proposed system improves delivery efficiency and enhances the quality of catering delivery services at UMKM Vfoodia.
Penerapan Metode Particle Swarm Optimization (PSO) untuk Optimasi Waktu Tunggu pada Sistem Pemesanan Jasa Servis Nur, Muhamad; Marisa, Marisa; Pratama, Fadhel Rizky
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.962

Abstract

In the competitive service industry, digital transformation of booking management systems has become essential for maintaining customer loyalty. However, many enterprises still rely on manual methods that result in high latency, queue congestion, and imbalanced technician workloads. This study aims to address these inefficiencies by implementing the Particle Swarm Optimization (PSO) algorithm within a web-based service booking system architecture. PSO, a metaheuristic algorithm inspired by the social behavior of animal swarms, is employed to search for globally optimal solutions in a multidimensional search space. The algorithm is configured with 20 particles, a maximum of 100 iterations, and parameters c1 = 2.0, c2 = 2.0, and w = 0.7 to minimize cumulative customer waiting time while balancing technician task allocation based on technician availability, service duration, and operational hour constraints (08:00–16:00). Empirical testing demonstrated significant improvements in operational performance. Prior to optimization, the total customer waiting time over a three-day observation period reached 380 minutes. Following PSO implementation, waiting time was drastically reduced to 150 minutes, representing a 60.53% reduction (230 minutes saved). These findings confirm that the PSO approach not only delivers rapid and adaptive solutions to real-time data fluctuations but also enhances operational system scalability. This research provides a practical contribution for service management system developers seeking to integrate computational intelligence into the optimization of complex business processes.
Leadership Support System for the Selection of Outstanding Students Using the MAUT Method Hidayat, Ahmad; Rahmayanti , Syifa Nurani; Amalya , Farida
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.983

Abstract

A student is a person who is currently in education. Outstanding students are students who are studying or studying in higher education, both in universities, as well as institutes or academies that are being followed in order to achieve their desired goals. They are people who are prospective intellectuals who are registered as students at a college or university called students. Based on the problems that often occur, a decision support system is needed to be used in solving these problems, so every student who is continuing their education is expected to study seriously for the sake of achieving achievements according to what is desired so that in the future they will become the next generation for the nation. In this study, 7 criteria were used in the selection of outstanding students that became a reference or assessment worthy of becoming outstanding students, including activeness, certificates obtained during lectures, cumulative achievement index scores, behavior in lectures, written works, the number of parental dependents, and income. Based on the problems that often occur, a decision support system is needed to be used in solving these problems, so every student who is continuing their education is expected to study seriously for the sake of achieving achievements according to what is desired so that in the future they will become the next generation for the nation. Therefore, a decision support system is needed to solve a problem by applying the Centroid Rank Order (ROC) Method used to carry out weighting and the Multi Attribute Utility Theory (MAUT) Method which is used to solve the problem by including weighting. From the test results, the best alternative that is considered worthy as an outstanding student is in  the A3 alternative  for Ira using  the Rank Order Centroid (ROC) Method and the Multi Attribute Utility Theory (MAUT) method with a score of 1.297.
Analisis Visual Perilaku Agen Q-Learning dan SARSA pada Cliff Walking Problem dengan Explainable Reinforcement Learning Atqiya, Firas; Sholahuddin, Muhammad Rizqi
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.984

Abstract

Reinforcement Learning (RL) has achieved remarkable success in complex sequential decision tasks. However, modern RL models often lack explainability, creating a serious "black box" problem, especially in high-stakes domains. This study proposes a Pygame-based real-time visualization architecture for RL, and demonstrates its benefits in a Cliff Walking case study using Q-Learning and SARSA algorithms. Key contributions include: (1) a real-time visualization architecture that decouples training logic from graphics rendering with support more than 60 FPS, (2) interpretive visualization techniques including diverging heatmaps, dynamic policy arrows, and Ghost Policies, and (3) a comprehensive empirical study clarifying the distinct characteristics of both algorithms. Experimental results clearly show that Q-Learning selects an efficient but risky path aligned with its optimistic off-policy nature, while SARSA converges on a safer path reflecting its on-policy nature that considers exploration safety. Quantitatively, Q-Learning successfully achieved an optimal 13-step path with an accumulation of 10,642 falls, whereas SARSA converged to a safe 23-step path with a significantly higher collision frequency (232,844 times) to avoid extreme penalties from the cliff zone.
Sistem Pendukung Keputusan Rekomendasi Calon Ketua BEM Pada Perguruan Tinggi Menggunakan Metode Weighted Product (WP) dan MOORA Syahputra, Irwan; Zalukhu, Anzas Ibezato; Hulu, Adil Priman Hati; Sartika, Dewi; Suhardiansyah, Suhardiansyah
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.992

Abstract

The BEM Chairperson is the highest position in an intra-campus organization elected by students. The process of selecting the BEM chairperson begins with a selection and choice based on several specific criteria that have different assessment weights. To determine the decision-making process, a decision support system (DSS) is one of the tools in solving this problem. Weight Product (WP) and MOORA are methods in SPK that are widely used to resolve decision-making that has many criteria and ranking systems. This study compares the WP and MOORA methods in selecting BEM chairperson candidates. Based on the calculation results, it is found that both methods can produce a sequence of BEM chairperson candidates. The stages of completing the WP and MOORA methods are determining the decision criteria, determining the weight of each criterion, determining alternatives and their values ??for each criterion, creating a decision matrix, normalizing the decision matrix, calculating the preference value for each alternative, sorting the largest value as a decision recommendation. Based on the results of the calculations for both methods, we can compare that the MOORA method, based on the Recommendation Assessment for the Election of Student Executive Board Chairperson at Budidarma University, Medan, selected alternative A1, Muhammad Aldi, S.Kom, with an optimization value of 0.4572. However, using the WP method, the selection of alternative A1, Muhammad Aldi, with a preference value of 0.193116, was selected as BEM Chairperson.
Sistem Prediksi Produksi Kelapa Sawit Berbasis Gradio Menggunakan Algoritma Regresi Linear Berganda Matondang, Irfan Jamal; Budianita, Elvia; Syafria, Fadhilah; Afrianty, Iis
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.994

Abstract

The instability of oil palm production often leads to discrepancies between production targets and actual outputs, thereby necessitating an accurate prediction model to support operational planning. This study aims to develop an oil palm production prediction model and to identify the most influential variables affecting production outcomes as a basis for data-driven decision-making. The model was developed using the Multiple Linear Regression method based on historical data from 2020–2024, consisting of 60 monthly observations with variables including number of trees, land area, rainfall, number of fruit bunches, and plant age. The research stages included data preprocessing, variable selection through testing several feature combinations, model development, and performance evaluation using the coefficient of determination (R²), Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Square Error (RMSE). The results indicate that the combination of number of trees, land area, number of fruit bunches, and plant age produced the best performance, with an R² value of 0.85 on the training data and 0.81 on the testing data. The MAE values were 125,307 kg and 176,984 kg, the MSE values were 28,870,838,455 kg² and 52,809,954,662 kg², and the RMSE values were 169,914 kg and 229,804 kg, respectively. Based on the regression coefficients, the number of fruit bunches was identified as the most dominant variable, with a coefficient value of 637,720 kg. The model was subsequently implemented using the Python Gradio library in the form of an interactive interface to support production planning effectiveness and minimize the risk of inaccurate decision-making in oil palm plantation management.

Page 1 of 2 | Total Record : 19

 1   2 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue Vol. 6 No. 2 (2026): February 2026 Vol. 6 No. 1 (2025): December 2025 Vol. 5 No. 6 (2025): October 2025 Vol. 5 No. 5 (2025): August 2025 Vol. 5 No. 4 (2025): June 2025 Vol. 5 No. 3 (2025): April 2025 Vol. 5 No. 2 (2025): February 2025 Vol. 5 No. 1 (2024): December 2024 Vol. 4 No. 6 (2024): Oktober 2024 Vol. 4 No. 5 (2024): August 2024 Vol. 4 No. 4 (2024): June 2024 Vol. 4 No. 3 (2024): April 2024 Vol. 4 No. 2 (2024): Februari 2024 Vol. 4 No. 1 (2023): December 2023 Vol. 3 No. 6 (2023): October 2023 Vol. 3 No. 5 (2023): Agustus 2023 Vol. 3 No. 4 (2023): June 2023 Vol. 3 No. 3 (2023): April 2023 Vol. 3 No. 2 (2023): Februari 2023 Vol. 3 No. 1 (2022): Desember 2022 Vol. 2 No. 3 (2022): Agustus 2022 Vol. 2 No. 2 (2022): April 2022 Vol. 2 No. 1 (2021): Desember 2021 Vol. 1 No. 3 (2021): Agustus 2021 Vol. 1 No. 2 (2021): April 2021 Vol. 1 No. 1 (2020): Desember 2020 More Issue