cover
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota malang,
Jawa timur
INDONESIA
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control
ISSN : 25032259     EISSN : 25032267     DOI : -
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control was published by Universitas Muhammadiyah Malang. journal is open access journal in the field of Informatics and Electrical Engineering. This journal is available for researchers who want to improve their knowledge in those particular areas and intended to spread the knowledge as the result of studies. KINETIK journal is a scientific research journal for Informatics and Electrical Engineering. It is open for anyone who desire to develop knowledge based on qualified research in any field. Submitted papers are evaluated by anonymous referees by double-blind peer review for contribution, originality, relevance, and presentation. The Editor shall inform you of the results of the review as soon as possible, hopefully within 4 - 8 weeks. The research article submitted to this online journal will be peer-reviewed at least 2 (two) reviewers. The accepted research articles will be available online following the journal peer-reviewing process.
Arjuna Subject : -
Articles 584 Documents
Hate Speech Analysis of YouTube Comments on the 2024 Indonesian Presidential Debate Using IndoBERT Agus Sasmito Aribowo; Yuli Fauziah; Yusna Bantulu; Shoffan Saifullah; Azfa Mutiara Ahmad Fubalo
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2604

Abstract

The rapid digitization of political campaigns has intensified the spread of hate speech on social media, threatening democratic discourse and social cohesion. In Indonesia, YouTube comments on the 2024 presidential election debates have emerged as a critical yet underexplored source of polarizing content. However, existing detection systems struggle with Indonesian-specific linguistic features, code-mixing, and implicit political sarcasm, while high annotation costs and class imbalance further limit the scalability of supervised approaches. To address these challenges, this study introduces a large-scale dataset of 38,742 YouTube comments collected from the five official debate stages and labeled using a cost-effective semi-supervised framework (20% expert-annotated, 80% pseudo-labeled). We systematically evaluate four classification models —IndoBERT, mBERT, SVM, and Random Forest—under identical experimental conditions using evaluation metrics optimized for imbalanced data. Experimental results demonstrate that IndoBERT consistently outperforms all baseline models, achieving an average accuracy of 89.7% and a macro F1-score of 0.89 across all debate stages. Notably, IndoBERT maintains high recall for the minority hate speech class (0.81–0.90), confirming its superior ability to capture localized political rhetoric and contextual nuances that multilingual and classical models frequently miss. This study contributes a publicly available Indonesian political hate speech dataset, validates a scalable semi-supervised annotation pipeline, and provides empirical evidence that domain-specific Transformer models are essential for reliable content moderation in politically charged, low-resource environments.
Comparison of Word2Vec and GloVe Performance in Bi-LSTM Models for Indonesian News Classification Muhammad Faris Wafda; Husni; Ika Oktavia Suzanti; Firdaus Solihin; Mula'ab; Army Justitia
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2608

Abstract

The explosion in the volume of textual data from digital news presents challenges in classifying content automatically and efficiently. For the task of classifying Indonesian-language news, this study aims to compare the performance of several word embeddings specifically Word2Vec using CBOW and Skip-Gram architectures and GloVe when applied to a Bidirectional Long Short-Term Memory (Bi-LSTM) model. This study uses a dataset consisting of 6,715 news articles from the Indonesian news portal that have undergone pre-processing, divided into five categories. The model was trained using 80% of the training data with K-Fold Cross Validation (K=5), while the remaining 20% of the data was used for testing. The experimental findings indicate that the Bi-LSTM model, when combined with CBOW embedding, yielded the best performance, achieving 95.16% accuracy and a 95.15% F1-Score. The Skip-Gram model followed with solid performance, achieving an accuracy of 93.30% and the fastest computation time. Conversely, the model that used pre-trained GloVe embedding delivered the poorest performance, achieving 88.98% accuracy. This result suggests that training embeddings on a specific domain is more effective at capturing local context. The conclusion of this study confirms that selecting a word embedding method specifically trained on local datasets is also an important step in achieving optimal accuracy in Indonesian news text classification.
A Comparative Study of Hybrid GARCH–HOLT–BPNN Models for Rainfall Forecasting Using a MATLAB-Based Intelligent Computing System Supardi Supardi; Syaharuddin Syaharuddin; Vera Mandailina; Saba Mehmood
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2636

Abstract

Rainfall forecasting is essential for water resource management, hydrometeorological disaster mitigation, and agricultural planning. This study addressed the limitations of previous research that focused on single models or hybrid approaches combining only two methods, which often failed to capture the simultaneous volatility, trend, and nonlinear characteristics of rainfall data. Monthly rainfall data from 2015 to 2024 were analyzed using three individual models: Generalized Autoregressive Conditional Heteroskedasticity (GARCH), Holt’s Exponential Smoothing, and Backpropagation Neural Network (BPNN). Two hybrid models, GARCH–Holt and GARCH–Holt–BPNN, were also developed to integrate the advantages of statistical and artificial intelligence methods. Hyperparameter tuning was performed to optimize model performance, and forecasting accuracy was evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). Results showed that GARCH effectively captured short-term volatility, Holt followed trends and seasonal patterns, and BPNN modeled nonlinear relationships despite sensitivity to data variations. The GARCH–Holt hybrid improved stability and accuracy compared to individual models, while the GARCH–Holt–BPNN hybrid achieved the highest predictive performance with a MAPE of 1.13%, indicating strong generalization capability. Forecasted rainfall for 2025 revealed seasonal patterns characterized by periods of heavy, moderate, and light rainfall. A MATLAB-based Graphical User Interface (GUI) was developed to facilitate interactive modeling and visualization. Overall, the proposed hybrid GARCH–Holt–BPNN model provides a more robust and reliable forecasting framework by effectively integrating volatility, trend, and nonlinear components, thereby enhancing predictive accuracy and supporting data-driven decision-making in hydrometeorological applications.
Automated Breast Cancer Cell Counting: Comparing Multi-class Segmentation and Two-stage Classification Strategies Dzaky Hanif Arjuna; Edy Kurniawan; Reza Fuad Rachmadi; I Ketut Eddy Purnama
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2639

Abstract

The manual interpretation of Hematoxylin and Eosin (H&E) histopathology images for breast cancer diagnosis is hindered by time constraints and observer bias. This research aims to develop an automated system using deep learning for cell detection and classification by evaluating two key approaches: Multi-class Segmentation (single-stage) and Segmentation followed by Classification (two-stage). The U-Net architecture was employed for segmentation, while MobileNetV2 and VGG16 were used for classification. The models were evaluated on the public IHC4BC dataset and primary data from Airlangga University Hospital (RSUA). The study also evaluated the impact of Resizing and Tiling data processing strategies. Experimental results showed that although the MobileNetV2 and VGG16 classification models achieved a high testing accuracy of 98.80%, the integrated two-stage system exhibited a high counting error, with a Mean Absolute Error (MAE) of 119.87 for positive cells, primarily due to under-segmentation of overlapping cells. In contrast, the Multi-class Segmentation approach utilizing the Tiling strategy demonstrated superior performance. This model effectively preserved spatial resolution while distinguishing cell types simultaneously, achieving the lowest MAE of 18.46 for positive cells and 1.66 for negative cells. This study concludes that Multi-class Segmentation with the Tiling strategy is the most effective and accurate approach for automated cell counting in histopathology images.
Scalable Multi-Agent Formation Control in RTS Games: A Virtual Anchor and Fluid-Based Allocation Ibnu Athaillah; Moch. Kholil
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2643

Abstract

The control system for troop formation movement is a critical component of Real-Time Strategy (RTS) games, directly affecting gameplay quality and player experience. However, traditional CPU-bound pathfinding algorithms fail to maintain 60 FPS when handling more than 1,000 agents in dynamic environments, particularly when balancing rigid formation structure with pathfinding efficiency amid complex obstacles. This study proposes an integrated framework for troop formation movement that synthesizes a virtual Anchor navigation paradigm with a Fluid-Based Formation Position Allocation algorithm.  Unlike traditional leader-follower methods, the proposed system utilizes a virtual anchor to calculate global pathfinding via NavMesh, while constituent agents dynamically adjust their positions relative to this reference point. To mitigate trajectory conflicts during formation changes, the system employs a fluid-dynamics-inspired sorting strategy that deterministically maps agents to target slots using parallel processing. The architecture is optimized for real-time performance using the Unity Job System, allowing for the coordination of large-scale agent aggregates. Experimental validation was conducted through behavioral scenarios—including Tunnel, Split, and Crowd tests—as well as stress tests involving up to 4,096 agents. The results demonstrate that the system successfully maintains formation integrity, executes autonomous regrouping after obstacle traversal, and ensures collision-free movement. Performance analysis further indicates that the control logic remains computationally stable at scale, with the primary limitations shifting to graphical rendering overhead rather than algorithmic complexity.
Kalman Filter-Based RSS Preprocessing for Cryptographic Key Generation in Zero-Knowledge Feige Fiat Shamir Authentication M. Cahyo Kriswantoro; Eko Handoyo; Ahmad Lathif Aditya
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2644

Abstract

Secure authentication in wireless communication environments requires mechanisms capable of verifying identity without exposing confidential information. Zero-Knowledge Authentication (ZKA) addresses this challenge by enabling interactive identity verification without revealing secret credentials. However, in practical wireless implementations, the reliability of ZKA strongly depends on the quality and consistency of the cryptographic keys used during the authentication process. One promising approach is to generate keys from physical-layer characteristics, such as Received Signal Strength (RSS). Nevertheless, raw RSS measurements are highly unstable due to noise, interference, mobility, and signal fluctuations, which often result in low reciprocity and key mismatch between legitimate nodes. This study proposes a Kalman Filter-based preprocessing method to improve RSS quality prior to cryptographic key generation for Zero-Knowledge Feige–Fiat–Shamir authentication. The Kalman Filter is employed to suppress measurement noise and enhance reciprocity between communicating nodes, allowing both parties to derive more consistent symmetric keys. The generated keys are then integrated into the authentication process to replace conventional static, channel-dependent key components. The proposed approach was evaluated using key consistency, entropy level, and bit mismatch rate as performance metrics. Experimental results show that Kalman Filter-based preprocessing significantly improves RSS stability and increases the correlation between legitimate nodes compared with unfiltered RSS measurements. In addition, the generated keys exhibit lower bit mismatch rates and better entropy characteristics, leading to higher authentication success rates while preserving the confidentiality properties of Zero-Knowledge Authentication. These findings demonstrate that Kalman Filter-assisted RSS preprocessing provides an effective and lightweight solution for strengthening cryptographic key generation and improving the reliability of authentication systems in wireless communication environments.
Poultry Disease Classification Using EfficientNetV2-L and MobileNetV2 Based on Fecal Images Rosida Vivin Nahari; Anisyafaah; Riza Alfita
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2648

Abstract

The timely identification of poultry diseases is essential for maintaining high livestock yields and curbing the transmission of infections, which directly supports global food security. The integration of multi-layered neural architectures has advanced this diagnostic process, offering an innovative approach to automated health monitoring through enhanced classification accuracy. By employing Convolutional Neural Networks (CNNs), the extraction of discriminative features from fecal images is automated, providing a scalable and robust approach to sustainable agriculture. This study proposes poultry disease classification using two CNN architectures, EfficientNetV2-L and MobileNetV2, trained under three scenarios: baseline, class weights, and Focal Loss. Using a dataset of 6,812 chicken fecal images, the experimental results demonstrate that applying Focal Loss significantly improves performance across all metrics. The EfficientNetV2-L model with Focal Loss achieved superior results, with 99.51% accuracy, 99.57% precision, 99.51% recall, and 99.52% F1-score. Meanwhile, MobileNetV2 performed reasonably well with a faster training time, making it suitable for resource-constrained environments. These findings indicate that combining Focal Loss with efficient CNN architectures enhances the classification of imbalanced datasets and provides a promising technological solution for real-time poultry disease detection systems.
Optimization of Retargeting Motion Capture for Remo Dance Using Fuzzy Logic Didit Prasetyo; Nugrahardi Ramadhani; Kartika Kusuma Wardani; Indriana Dwi Andiany
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2663

Abstract

Retargeting motion capture for traditional dance animation faces challenges in maintaining biomechanical accuracy while preserving cultural expressiveness, especially when human motion data are transferred to character models with different skeletal structures. This research aims to optimize the retargeting of East Java Remo dance through an adaptive artificial intelligence-based evaluation approach. The Remo dance movement was recorded using a multi-camera optical motion capture system and retargeted to two types of 3D characters: realistic and stylized. The evaluation was conducted using quantitative metrics (Mean Squared Error, Structural Similarity Index, Dynamic Time Warping, and Kalman Filtering), as well as a qualitative approach through Laban Movement Analysis. Subsequently, Mamdani fuzzy logic was integrated to synthesize all these parameters into the Fuzzy Retargeting Quality Score (FRQS). The results showed that the realistic character had higher movement accuracy (MSE = 0.0032; SSI = 0.89; DTW = 0.92) and obtained an FRQS value of 86.4 (very optimal category), whereas the stylized character obtained an FRQS of 71.2 (moderately optimal), reflecting a compromise between movement precision and visual appeal. The integration of fuzzy logic allows for more contextual and human-centric retargeting evaluation, as well as strengthening the dual-model approach to preserving and educating about traditional dance through digital animation.
Federated Ensemble Learning with SHAP–LIME Interpretability for Smart Home Energy Prediction Rahma Puspitasari; Muhammad Arif Hermawan; Joshua Andrian; Siti Sendari; Ira Kumala Sari
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2665

Abstract

The increased adoption of IoT-based Smart Home systems in Indonesia has resulted in a growing volume of device-level energy data, opening up opportunities for the development of predictive models to support efficient household electricity consumption. However, challenges related to accuracy, interpretability, and data privacy remain a major concern, especially when data are distributed across multiple devices. This study evaluates the performance of four tree-based ensemble models, namely Random Forest, Gradient Boosting, XGBoost, and LightGBM, in centralized learning and federated learning scenarios using the Indonesia Smart Home Dataset. After undergoing feature preprocessing and refinement, including the removal of Sofa Pressure and Bed Pressure due to high noise, each model was trained and evaluated using MAE, MSE, and RMSE metrics. Federated learning was implemented through the Federated Averaging (FedAvg) algorithm to maintain data privacy without the need to transfer raw data between devices. The results show that LightGBM consistently provides the best performance in both scenarios and demonstrates resilience to data fragmentation and heterogeneity. Although there was a slight increase in error in federated learning, the error values remained within an acceptable range. SHAP and LIME analyses revealed that high-power devices such as air conditioners, water pumps, rice cookers, lights, and refrigerators had the greatest contribution.
Predicting Social Media Post Engagement and Virality Using Graph Neural Network Approaches and Content-based Features Fathimah Az Zahrah; Riska Dhenabayu; Muhammad Fajar Wahyudi Rahman; Renny Sari Dewi; Zamabhungane Hadebe Aminah
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 3, August 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i3.2686

Abstract

Social media teams increasingly rely on early signals to prioritize content, yet forecasting engagement and identifying viral posts remain difficult under temporal drift and heavy-tailed interaction counts. This study evaluated Graph Neural Network (GNN) approaches for predicting post engagement and virality from pre-posting content-based and contextual features. The Social Media Engagement Report dataset, which contained 100,000 posts across Twitter, LinkedIn, Facebook, and Instagram spanning March 2021–March 2024, was used. Post-release variables (impressions, reach, engagement rate) were excluded to prevent leakage. A homogeneous post–post graph was constructed using k-nearest-neighbor similarity in an embedding space and exact-match links on low-cardinality context. Ridge/Logistic Regression, Random Forest, and XGBoost as the baselines were compared against GraphSAGE and GAT under a chronological train, validation, and test split. Regression used MAE, RMSE, and R², while virality classification used ROC-AUC, PR-AUC, and Precision at the top 1% ranked posts. GraphSAGE yielded the strongest virality screening, achieving ROC-AUC = 0.66, PR-AUC = 0.54–0.56, and Precision@1% up to 0.75, substantially above non-graph baselines. For regression, GAT produced the lowest errors despite a negative R², indicating limited explained variance. Overall, similarity-graph GNNs are most effective for early virality identification, whereas exact count prediction remains challenging in a strictly pre-posting, time-aware setting. These findings contribute empirical evidence that, under leakage-safe pre-posting conditions, graph-based relational modeling is more useful for ranking and screening potentially viral posts than for precise engagement-count forecasting.

Filter by Year

2016 2026


Filter By Issues
All Issue Vol. 11, No. 4, November 2026 (Article in Progress) Vol. 11, No. 3, August 2026 Vol. 11, No. 2, May 2026 Vol. 11, No. 1, February 2026 Vol. 10, No. 4, November 2025 Vol. 10, No. 3, August 2025 Vol. 10, No. 2, May 2025 Vol. 10, No. 1, February 2025 Vol. 9, No. 4, November 2024 Vol. 9, No. 3, August 2024 Vol. 9, No. 2, May 2024 Vol. 9, No. 1, February 2024 Vol. 8, No. 4, November 2023 Vol. 8, No. 3, August 2023 Vol. 8, No. 2, May 2023 Vol. 8, No. 1, February 2023 Vol. 7, No. 4, November 2022 Vol. 7, No. 3, August 2022 Vol. 7, No. 2, May 2022 Vol. 7, No. 1, February 2022 Vol. 6, No. 4, November 2021 Vo. 6, No. 3, August 2021 Vol. 6, No. 2, May 2021 Vol. 6, No. 1, February 2021 Vol. 5, No. 4, November 2020 Vol. 5, No. 3, August 2020 Vol. 5, No. 2, May 2020 Vol. 5, No. 1, February 2020 Vol 4, No 4, November 2019 Vol 4, No 3, August 2019 Vol 4, No 2, May 2019 Vol 4, No 2, May 2019 Vol 4, No 1, February 2019 Vol 4, No 1, February 2019 Vol 3, No 4, November 2018 Vol 3, No 4, November 2018 Vol 3, No 3, August 2018 Vol 3, No 3, August 2018 Vol 3, No 2, May-2018 Vol 3, No 2, May-2018 Vol 3, No 1, February-2018 Vol 3, No 1, February-2018 Vol 2, No 4, November-2017 Vol 2, No 4, November-2017 Vol 2, No 3, August-2017 Vol 2, No 3, August-2017 Vol 2, No 2, May-2017 Vol 2, No 2, May-2017 Vol 2, No 1, February-2017 Vol 2, No 1, February-2017 Vol 1, No 3, November-2016 Vol 1, No 3, November-2016 Vol 1, No 2, August-2016 Vol 1, No 2, August-2016 Vol 1, No 1, May-2016 Vol 1, No 1, May-2016 More Issue