cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Agus Harjoko
Contact Email
ijccs.mipa@ugm.ac.id
Phone
+62274 555133
Journal Mail Official
ijccs.mipa@ugm.ac.id
Editorial Address
Gedung S1 Ruang 416 FMIPA UGM, Sekip Utara, Yogyakarta 55281
Location
Kab. sleman,
Daerah istimewa yogyakarta
INDONESIA
IJCCS (Indonesian Journal of Computing and Cybernetics Systems)
Published by Universitas Gadjah Mada
ISSN : 19781520     EISSN : 24607258     DOI : https://doi.org/10.22146/ijccs
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering
Indonesian Journal of Computing and Cybernetics Systems (IJCCS), a two times annually provides a forum for the full range of scholarly study . IJCCS focuses on advanced computational intelligence, including the synergetic integration of neural networks, fuzzy logic and eveolutionary computation, so that more intelligent system can be built to industrial applications. The topics include but not limited to : fuzzy logic, neural network, genetic algorithm and evolutionary computation, hybrid systems, adaptation and learning systems, distributed intelligence systems, network systems, human interface, biologically inspired evolutionary system, artificial life and industrial applications. The paper published in this journal implies that the work described has not been, and will not be published elsewhere, except in abstract, as part of a lecture, review or academic thesis.
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 516 Documents
 48   49   50   51   52 
Web-Based Laundry Customer Segmentation Using Agglomerative Hierarchical Clustering Halena Diyana Sedar; Friska Abadi; Irwan Budiman; Fatma Indriani; Setyo Wahyu Saputro
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.123059

Abstract

Within many small and medium-sized enterprises (SMEs) in the laundry sector, customer transaction data are predominantly utilized merely as historical records rather than being optimally processed for strategic business decision-making. Consequently, the potential of this data to elucidate customer behavioral patterns and inform targeted strategies remains largely untapped. This study aims to develop a customer segmentation feature for the Rayhan Laundry web application using Agglomerative Hierarchical Clustering (AHC). The method involved preprocessing transaction data, comprising customer identities, dates, and weights by generating unique IDs, handling missing values, and aggregating data based on total weight and transaction frequency. Utilizing Ward linkage and Euclidean distance, the AHC algorithm successfully partitioned customers into three distinct segments: high-volume, regular, and sporadic. The model achieved a silhouette score of 0.753, demonstrating robust cluster separation. The feature was seamlessly integrated into a Laravel-based web application, equipped with advanced search, filtering, pagination, and export functionalities. Functional validation via Blackbox Testing confirmed all features met specifications. Furthermore, usability evaluation using the System Usability Scale (SUS) yielded an average score of 82.75, categorized as “Excellent.” Conclusively, the developed feature is highly effective and user-friendly, providing a robust tool for data-driven customer relationship management in the laundry service industry.
Aspect-Based Sentiment Analysis of BRImo and Livin’ by Mandiri Reviews Yuli Syafitri
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.123518

Abstract

The rapid adoption of digital banking generates a large volume of unstructured user reviews, making document-level sentiment analysis insufficient to identify the specific sources of satisfaction and dissatisfaction.This study applies Aspect-Based Sentiment Analysis (ABSA) to user reviews of BRImo and Livin’ by Mandiri by comparing TF-IDF and Word2Vec representations using Support Vector Machine (SVM). Reviews were collected from Google Play Store with google-play-scraper using newest-first ordering and a target of up to 30,000 reviews per application.After cleaning and selection, 41,461 clean reviews remained, and sentiment classes were balanced to 8,943 records for modeling. Aspect extraction used a dictionary-based approach, while three sentiment classes were assigned from star ratings.TF-IDF used 1–2 grams with a maximum of 5,000 features, whereas Word2Vec used a 100-dimensional Skip-gram model with average pooling.SVM hyperparameters were optimized using Grid Search and 5-Fold Cross-Validation. Both feature schemes achieved 64.6% accuracy, 66.7% macro precision, 64.6% macro recall, and 65.3% macro F1-score. BRImo was dominated by positive sentiment (64.1%), whereas Livin’ by Mandiri was dominated by negative sentiment (49.2%), with Login as the most prominent issue. The findings indicate that TF-IDF and Word2Vec were practically equivalent on this dataset,while ABSA provides actionable evidence for prioritizing digital banking service improvements
Target Representation as a Key Factor for Generalization of Physics-Based Electricity Price Correction Models Istifa Shania Putri
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.123700

Abstract

Physics-based electricity price forecasting models, such as Linear Optimal Power Flow (LOPF), provide strong theoretical accuracy but often fail to capture real market prices. To reduce this gap, machine learning-based correction models are commonly applied; however, their reliability under significant market regime shifts remains unclear. This study investigates the robustness boundaries of price correction models using German electricity market data from 2019 and 2024, consisting of 17,544 hourly observations. Three key findings are identified. First, model selection and hyperparameter optimization do not improve cross-regime robustness: all 16 tested configurations fail when models trained in high-price regimes are applied to low-price regimes, increasing forecasting errors by 2.2–4.0 times. Second, the main source of failure is the target formulation. For the best-performing configuration, replacing the conventional price-difference target (EUR/MWh) with a ratio-based target turns all four transfer scenarios positive, achieving up to a 34.1% improvement in backward transfer. Third, model degradation is primarily driven by the distance between fuel-cost market regimes rather than temporal distance. Evaluation across 30 regime transitions shows that conventional correction approaches deteriorate as market condition differences increase. These findings reveal the limitations of current AI-based electricity price correction frameworks and highlight target redesign and regime-aware modeling as critical strategies for improving robustness under evolving market conditions.This study tests model resilience during the 2022 extreme market volatility (peak of the European gas crisis, RMSE: 159.06 EUR/MWh). The results show that the ratio target outperforms the difference target by +8% to +11% when predicting crisis periods. However, the model fails when trained on crisis data but tested under normal conditions, showing an explicit regime limit.
Comparison of Random Forest and XGBoost for Impulsive Buying Prediction Adinda Bintang Oktavia; Andy Prasetyo Utomo; Soni Adiyono
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.125433

Abstract

This study compares Random Forest and XGBoost for predicting impulsive buying among mobile marketplace users using historical data from 2022 and questionnaire data from 2026. The research stages include preprocessing, feature and target construction, stratified data splitting, class imbalance handling using SMOTE within a pipeline, and model evaluation using Accuracy, Precision, Recall, F1-Score, and Area Under the Curve (AUC). GridSearchCV with five-fold cross-validation was used for model selection, while SHAP was employed to interpret feature contributions to predictions. On the 2022 historical data, Random Forest achieved an F1-Score of 0.8129 and an AUC of 0.8260, outperforming XGBoost, which obtained an F1-Score of 0.7922 and an AUC of 0.7933. On the 2026 questionnaire data, both algorithms achieved an F1-Score of 0.9279, while Random Forest obtained an AUC of 0.8724 compared with 0.8432 for XGBoost. SHAP analysis identified Hedonic Browsing as the feature with the greatest predictive contribution in both datasets. Overall, Random Forest achieved a higher mean cross-validation F1-Score on both datasets.
Learning Analytics for STEM Course-Completion: Logistic Regression, ANN, and DNN Mihuandayani Mihuandayani; Michel Farrel Tomatala; Ritham Tuntun; Caroline Belinda Tangka
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.125791

Abstract

Identifying students at risk of not completing STEM courses is essential for timely academic intervention, yet institutional learning data remain underutilized. This study develops a comparative learning analytics framework for classifying STEM course completion using Logistic Regression, a Shallow Artificial Neural Network (ANN), and a Deep Neural Network (DNN). The dataset contained 1,800 records from 300 students in the 2022 and 2023 cohorts, covering six STEM courses and 14 predictors. Data were stratified into training, validation, and testing subsets of 70%, 15%, and 15%, with random oversampling applied only to the training set. Performance was evaluated using accuracy, precision, recall, F1-score, AUC, and confusion matrices, with non-completion as the positive class. Logistic Regression achieved 97.78% accuracy, 94.57% precision, 98.86% recall, 96.67% F1-score, and 99.93% AUC. Shallow ANN achieved 97.41% accuracy and 96.09% F1-score, while DNN achieved 97.78% accuracy, 95.56% precision, and 96.63% F1-score. Final-exam, midterm, assignment, and quiz scores were the most influential predictors. These findings indicate that Logistic Regression provides the strongest overall balance between predictive performance and interpretability for STEM learning analytics.
K-Means Retail Product Sales Clustering via Elbow and Davies-Bouldin Ni Wayan Wardani
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 20, No 4 (2026): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.125921

Abstract

 The expansion of the retail business requires business operators to accurately group best-selling and slow-moving products in order to optimize inventory adjustment strategies. This study aims to cluster sales transaction data and determine the optimal number of clusters using the K-Means algorithm. A secondary dataset consisting of 7,373 transaction records was analyzed based on the attributes of transaction count and total sales. The best cluster number was determined using the Elbow method based on the Sum of Squared Errors and validated using the Davies-Bouldin Index. The test results show that both methods consistently produced an optimal number of clusters, k, equal to 4, with a Davies-Bouldin Index value of 0.47. This value indicates a very well-separated cluster structure. The final clustering resulted in four product performance categories: 2 “best-selling” products, 20 “selling well” products, 171 “moderately selling” products, and 7,180 “poorly selling” products. This evaluation demonstrates that the combination of K-Means, the Elbow method, and the Davies-Bouldin Index effectively provides data-driven insights to support inventory management decision-making in the retail business.

Page 52 of 52 | Total Record : 516

 48   49   50   51   52 

Filter by Year

2006 2026


Reset
Filter By Issues
All Issue Vol 20, No 4 (2026): October Vol 20, No 3 (2026): July Vol 20, No 2 (2026): April Vol 20, No 1 (2026): January Vol 19, No 4 (2025): October Vol 19, No 3 (2025): July Vol 19, No 2 (2025): April Vol 19, No 1 (2025): January Vol 18, No 4 (2024): October Vol 18, No 3 (2024): July Vol 18, No 2 (2024): April Vol 18, No 1 (2024): January Vol 17, No 4 (2023): October Vol 17, No 3 (2023): July Vol 17, No 2 (2023): April Vol 17, No 1 (2023): January Vol 16, No 4 (2022): October Vol 16, No 3 (2022): July Vol 16, No 2 (2022): April Vol 16, No 1 (2022): January Vol 15, No 4 (2021): October Vol 15, No 3 (2021): July Vol 15, No 2 (2021): April Vol 15, No 1 (2021): January Vol 14, No 4 (2020): October Vol 14, No 3 (2020): July Vol 14, No 2 (2020): April Vol 14, No 1 (2020): January Vol 13, No 4 (2019): October Vol 13, No 3 (2019): July Vol 13, No 2 (2019): April Vol 13, No 1 (2019): January Vol 12, No 2 (2018): July Vol 12, No 1 (2018): January Vol 11, No 2 (2017): July Vol 11, No 1 (2017): January Vol 10, No 2 (2016): July Vol 10, No 1 (2016): January Vol 9, No 2 (2015): July Vol 9, No 1 (2015): January Vol 8, No 2 (2014): July Vol 8, No 1 (2014): January Vol 7, No 2 (2013): July Vol 7, No 1 (2013): January Vol 6, No 2 (2012): July Vol 6, No 1 (2012): January Vol 5, No 3 (2011): November Vol 5, No 2 (2011): July Vol 5, No 1 (2011): January Vol 1, No 2 (2007): July Vol 1, No 1 (2006): January More Issue