Jurnal Teknik Informatika (JUTIF)
Jurnal Teknik Informatika (JUTIF) is an Indonesian national journal, publishes high-quality research papers in the broad field of Informatics, Information Systems and Computer Science, which encompasses software engineering, information system development, computer systems, computer network, algorithms and computation, and social impact of information and telecommunication technology. Jurnal Teknik Informatika (JUTIF) is published by Informatics Department, Universitas Jenderal Soedirman twice a year, in June and December. All submissions are double-blind reviewed by peer reviewers. All papers must be submitted in BAHASA INDONESIA. JUTIF has P-ISSN : 2723-3863 and E-ISSN : 2723-3871. The journal accepts scientific research articles, review articles, and final project reports from the following fields : Computer systems organization : Computer architecture, embedded system, real-time computing 1. Networks : Network architecture, network protocol, network components, network performance evaluation, network service 2. Security : Cryptography, security services, intrusion detection system, hardware security, network security, information security, application security 3. Software organization : Interpreter, Middleware, Virtual machine, Operating system, Software quality 4. Software notations and tools : Programming paradigm, Programming language, Domain-specific language, Modeling language, Software framework, Integrated development environment 5. Software development : Software development process, Requirements analysis, Software design, Software construction, Software deployment, Software maintenance, Programming team, Open-source model 6. Theory of computation : Model of computation, Computational complexity 7. Algorithms : Algorithm design, Analysis of algorithms 8. Mathematics of computing : Discrete mathematics, Mathematical software, Information theory 9. Information systems : Database management system, Information storage systems, Enterprise information system, Social information systems, Geographic information system, Decision support system, Process control system, Multimedia information system, Data mining, Digital library, Computing platform, Digital marketing, World Wide Web, Information retrieval Human-computer interaction, Interaction design, Social computing, Ubiquitous computing, Visualization, Accessibility 10. Concurrency : Concurrent computing, Parallel computing, Distributed computing 11. Artificial intelligence : Natural language processing, Knowledge representation and reasoning, Computer vision, Automated planning and scheduling, Search methodology, Control method, Philosophy of artificial intelligence, Distributed artificial intelligence 12. Machine learning : Supervised learning, Unsupervised learning, Reinforcement learning, Multi-task learning 13. Graphics : Animation, Rendering, Image manipulation, Graphics processing unit, Mixed reality, Virtual reality, Image compression, Solid modeling 14. Applied computing : E-commerce, Enterprise software, Electronic publishing, Cyberwarfare, Electronic voting, Video game, Word processing, Operations research, Educational technology, Document management.
Articles
1,242 Documents
<![CDATA[Rainfall Prediction Using Feature Engineering, SMOTE, and Random Forest-XGBoost Soft Voting Ensemble on WeatherAUS Dataset ]]>
<![CDATA[Fareza Ahmad Kurniawan]]>;
<![CDATA[M. Faris Al Hakim]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.5974
<![CDATA[Rainfall prediction remains a critical challenge in meteorological science due to the non-linear and non-stationary nature of weather data. A persistent obstacle in building accurate rainfall classifiers is class imbalance, where no-rain observations significantly outnumber rain events, causing models to underperform on the minority Rain class precisely the class of greatest practical importance for flood preparedness and agricultural management. This study proposes an integrated machine learning pipeline combining feature engineering, Synthetic Minority Over-sampling Technique (SMOTE), Random Forest-based feature selection, hyperparameter tuning via Randomized Search Cross-Validation, and a soft voting ensemble of Random Forest and XGBoost for rainfall prediction. The WeatherAUS dataset containing 142,193 daily observations from 49 cities across Australia. Experimental results demonstrate that the baseline soft voting ensemble achieved the best overall performance with an Accuracy of 0.8498, ROC-AUC of 0.8800, and Weighted F1-Score of 0.8467, outperforming both standalone Random Forest and XGBoost across all metrics. Furthermore, the study finds that preprocessing quality — specifically the integration of feature engineering, SMOTE balancing, and feature selection — has a greater influence on model performance than hyperparameter optimization. This research contributes an evidence-based integrated preprocessing framework for imbalanced meteorological classification that advances the application of ensemble machine learning in rainfall prediction, with practical implications for early warning systems and data-driven weather forecasting in the field of Informatics and Computer Science.]]>
<![CDATA[Enhanced BiLSTM-CNN with Adaptive Regularization for Sequential Lung Cancer CT Scan Segmentation ]]>
<![CDATA[Juanda Hakim Lubis]]>;
<![CDATA[Agus Perdana Windarto]]>;
<![CDATA[Sundari Retno Andani]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.5999
<![CDATA[Lung cancer remains one of the leading causes of cancer-related mortality worldwide, making accurate and consistent CT scan segmentation essential for early diagnosis and treatment planning. This study proposes an enhanced Bidirectional Long Short-Term Memory-Convolutional Neural Network (BiLSTM-CNN) framework for sequential lung cancer CT scan segmentation. The proposed model integrates stacked BiLSTM layers, dropout regularization, adaptive learning-rate scheduling, and an Enhanced ResNet-18 backbone to improve temporal consistency and spatial feature representation across sequential CT slices. Unlike conventional CNN-based segmentation approaches that process slices independently, the proposed framework captures inter-slice contextual dependencies to produce more stable and anatomically consistent segmentation results. The model was evaluated using a publicly available lung cancer CT scan dataset with an 80:20 training–testing split. Experimental results demonstrate that the proposed method outperformed standard LSTM and conventional BiLSTM models, achieving superior segmentation performance with a Dice Similarity Coefficient (DSC) of 0.6960 and an Intersection over Union (IoU) of 0.5337. The findings indicate that the integration of bidirectional temporal modeling and enhanced feature extraction significantly improves segmentation accuracy and generalization capability. This study contributes to the development of lightweight and efficient AI-based medical imaging systems that support more reliable lung cancer diagnosis and clinical decision-making.]]>
<![CDATA[Particle Swarm Optimization for Hyperparameter Tuning in FedProx-Based Federated Learning Using DenseNet-201 for Breast Cancer Classification ]]>
<![CDATA[Rizal Dwi Anggoro]]>;
<![CDATA[Winarno Winarno]]>;
<![CDATA[Ery Permana Yudha]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.6012
<![CDATA[Breast cancer remains one of the leading causes of cancer-related mortality among women worldwide, making early detection a critical clinical priority. While deep learning has demonstrated strong performance in mammogram classification, its development is constrained by data privacy regulations that prevent centralized data collection across medical institutions. Federated learning (FL) offers a promising solution by enabling distributed model training without transferring raw patient data. However, FL faces significant challenges under Non-Independent and Identically Distributed (Non-IID) data conditions, which are common in real-world medical settings and can degrade model performance and convergence stability. This study proposes a breast cancer classification system based on FL using FedProx and DenseNet-201, with hyperparameter optimization via Particle Swarm Optimization (PSO) to improve model performance under heterogeneous data distributions. Experiments were conducted across multiple Dirichlet distribution scenarios (α = 0.1, 0.3, 0.5) comparing FL baseline and FL PSO based. Results show that PSO consistently improved performance across all scenarios, with the most significant gain observed under highly Non-IID conditions (α = 0.1), where the F1-score increased from 83.18% to 88.00%. PSO-optimized FL also demonstrated faster convergence, reducing the number of rounds required to reach optimal performance. Furthermore, the optimized configuration yielded lower training time per round compared to the baseline. These findings indicate that PSO-based hyperparameter optimization effectively enhances FL performance under data heterogeneity while preserving patient data privacy.]]>
<![CDATA[Semantic-Preserving Hybrid Tokenization With Codebert For Robust Webshell Detection In Hypertext Preprocessor Source Code ]]>
<![CDATA[Muhammad Kevin Adli Pratama]]>;
<![CDATA[Muh Ghazy Daffa Sampe]]>;
<![CDATA[Muhammad Ariando Ferdian]]>;
<![CDATA[Putri Tendry Zahrany]]>;
<![CDATA[Alif Rifa'i]]>;
<![CDATA[Anindita Septiarini]]>;
<![CDATA[Joan Angelina Widians]]>;
<![CDATA[Akhmad Irsyad]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.6121
<![CDATA[Transformer-based code models can achieve high accuracy in PHP(Hypertext Preprocessor) webshell detection, but high in-distribution scores do not necessarily indicate reliable behavior under difficult benign cases. This study investigates shortcut learning in CodeBERT-based PHP webshell classification and proposes a semantic-preserving hybrid representation that combines PHP structural tokens with explicit security-relevant tokens. After curation and SHA-256 exact deduplication, the corpus consists of 6,430 PHP files, including 3,685 benign files and 2,745 malicious webshell samples. The proposed tokenizer normalizes remote code execution sinks, input sources, obfuscation functions, and file operations into explicit tokens such as SINK_RCE_SYSTEM, INPUT_POST, OBF_BASE64, and FILE_INCLUDE. CodeBERT is fine-tuned under repository-aware and cross-partition robustness settings and evaluated using global metrics, confusion matrices, false-positive rates by benign bucket, and attention-based interpretation. Hybrid Tokens achieved an accuracy of 0.9902 and an F1-score of 0.9886 in repository-aware testing, while maintaining an accuracy of 0.9695 and an F1-score of 0.9629 in cross-partition robustness testing. The results indicate that preserving security semantics reduces reliance on superficial artifacts while retaining behavior-critical cues for distinguishing procedural benign scripts from webshells. These findings reframe PHP webshell detection as a robustness problem rather than a pure accuracy-maximization task, and show that security-aware input representation is a practical lever for reducing shortcut dependence. The study contributes to robust code intelligence, security-oriented representation learning, and more reliable evaluation of machine-learning detectors for cybersecurity.]]>
<![CDATA[Software Development Cost Prediction Using XGBoost with Optuna-Based Hyperparameter Optimization on the COCOMO Dataset ]]>
<![CDATA[Eddy Maryanto]]>;
<![CDATA[Bangun Wijayanto]]>;
<![CDATA[Swahesti Puspita Rahayu]]>;
<![CDATA[Dwi Kurnia Wibowo]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 3 (2026): JUTIF Volume 7, Number 3, June 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.3.6361
<![CDATA[Software development cost estimation remains a major challenge in project management because the complexity of software projects and the uncertainty of cost-influencing factors often lead to inaccurate estimates, adversely affecting resource allocation, budgeting, and project planning. This study aims to develop and evaluate an XGBoost-based software development cost prediction model optimized through Optuna-based hyperparameter optimization, as well as to investigate whether hyperparameter optimization can significantly improve prediction accuracy compared with the default XGBoost configuration. The proposed approach was evaluated using the Constructive Cost Model (COCOMO) dataset, a widely used benchmark for software cost estimation. The methodology consists of data preprocessing, XGBoost model training, Optuna-based hyperparameter optimization, and model evaluation using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Magnitude of Relative Error (MMRE), and the coefficient of determination (R²). Experimental results show that the optimized XGBoost model significantly outperformed the default configuration. The default model achieved an MAE of 189.6093, an RMSE of 296.1262, an MMRE of 132.42%, and an R² of 0.7162, whereas the optimized model reduced the MAE to 101.4871, RMSE to 134.7717, and MMRE to 34.04%, while increasing the R² to 0.9412. These results demonstrate that Optuna-based hyperparameter optimization substantially enhances the predictive performance of XGBoost for software cost estimation. The proposed approach provides a reliable decision-support tool for software project managers, enabling more accurate cost estimation, improved resource planning, and more effective project management.]]>
<![CDATA[Improvement Deep Learning Model for Batik Classification Used EfficientNetB2 and Augmentation Data ]]>
<![CDATA[Ismi Kusumaningroem]]>;
<![CDATA[M.Arief Soeleman]]>;
<![CDATA[Fikri Budiman]]>;
<![CDATA[M. Hasan Nafi]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.4902
<![CDATA[Batik is one of Indonesia’s cultural heritages that has been recognized by UNESCO. The diversity of batik motifs in Indonesia presents a challenge in classifying each motif from different regions. The purpose of this research is to improve batik motif classification using a deep learning model with the Convolutional Neural Network EfficientNetB2 architecture combined with data augmentation techniques. This study uses a dataset from Kaggle consisting of 1,964 images with seven batik motifs from the northern coast of Java: buketan, jlamprang, liong, mega mendung, negative, singa barong, and tujuh rupa. The augmentation technique applied in this study is traditional augmentation, including transformations such as horizontal flipping, rotation, and brightness adjustment. This was implemented to overcome dataset limitations, reduce the risk of overfitting, and increase dataset variation. The model integrates transfer learning through ImageNet pre-training to optimize the modification of the EfficientNetB2 architecture for specific batik motifs. EfficientNetB2 achieved an accuracy of 99% on the validation data, with precision, recall, and F1-scores consistently above 98%. This result is higher compared to previous research that applied the MPSO technique, which achieved 94% accuracy with precision, recall, and F1-scores of 78%. Model performance evaluation was conducted using a confusion matrix. Experimental results show that minimal misclassification occurred on the negative motif, with three samples incorrectly classified as jlamprang. These findings indicate that EfficientNetB2 combined with data augmentation significantly improves classification accuracy. Moreover, the use of the EfficientNetB2 architecture does not require long computational time, making it highly practical and efficient. This research contributes to the field of machine learning, the preservation of batik culture, and has the potential to support the batik industry in the commercial sector.]]>
<![CDATA[An Essence-Based Software Development Methodology to Support a Hybrid Waterfall-Agile in the Banking Domain: A Case Study at Bank XYZ ]]>
<![CDATA[Muhammad Ali Ihsan Fauzi]]>;
<![CDATA[Eko K. Budiardjo]]>;
<![CDATA[Heru Martin Saputra]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.5613
<![CDATA[Bank XYZ experienced a decline in on-time software project delivery, from 93.02% in 2021 to 85.00% in 2023, indicating that its predominantly sequential Waterfall process was insufficiently adaptive to changing requirements, while partial agile adoption remained unstandardized across teams. This study aims to design an Essence Framework-based hybrid Waterfall-Agile methodology that preserves governance, documentation, and approval controls required in a regulated banking environment while enabling faster iteration and better work visibility. The study employed a Design Science Research approach. Problem diagnosis was conducted through interviews and document analysis, while literature synthesis was used to build a practice backlog. Selected practices were mapped to Essence elements and assembled into a five-stage methodology consisting of Planning, Designing & Development, Testing, Release & Deployment, and Post-Implementation, followed by a pilot implementation in one project and qualitative evaluation through semi-structured interviews with developers, QA members, and external experts. The results show that the proposed methodology improved work visibility through Kanban-based flow visualization, increased clarity and traceability of workflow and artefacts, and remained fully compatible with the existing stage-gate and regulatory controls. However, the evaluation also identified gaps in layered testing, particularly unit testing and test automation, along with the need for clearer transition guidance. These findings contribute empirical evidence that Essence can support traceable and structured tailoring of hybrid software development methodologies in high-compliance financial environments, thereby strengthening software engineering process maturity in regulated sectors]]>
<![CDATA[Comparing Recommender System Algorithms: From Traditional Collaborative Filtering to Deep Learning Autoencoder on MovieLens Datasets ]]>
<![CDATA[Imam Fahrur Rozi]]>;
<![CDATA[Triyanna Widiyaningtyas]]>;
<![CDATA[Wahyu Caesarendra]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.5710
<![CDATA[With the increasing prevalence of digital platforms, recommender systems play a crucial role in personalizing user experiences. However, selecting an appropriate recommender system algorithm remains challenging due to inconsistent evaluation procedures and limited comparative studies that simultaneously address accuracy, efficiency, and architectural extensibility. This study compares nine recommender system algorithms, including traditional collaborative filtering (SVD, SVD++, KNN, NMF, SlopeOne), content-based filtering, hybrid approach, and deep learning autoencoder technique using MovieLens-1M and MovieLens-100K dataset. The hybrid approach achieved the best accuracy (RMSE 0.6865), followed by SVD++ and SVD (RMSE 0.8660 and 0.8729). Autoencoder demonstrated moderate performance (RMSE 0.9961) with 14% accuracy gap from SVD. Computational efficiency varied greatly, from 0.29 seconds for content-based methods to 3,470.73 seconds for SVD++. Cross-dataset validation of collaborative filtering methods confirmed consistency across datasets. Beyond accuracy metrics, architectural analysis identified autoencoder's extensibility advantage for integrating heterogeneous data sources, an important consideration for systems requiring multimodal capabilities. The findings provide empirical insights for method selection based on system priorities such as accuracy, efficiency, and architectural flexibility for multimodal integration.]]>
<![CDATA[Optimization of Drone Routing Problem with Energy Constraints and Charging Station Integration using Improved ParthenoGenetic Algorithm ]]>
<![CDATA[Sugiarto Cokrowibowo]]>;
<![CDATA[A. Amirul Asnan Cirua]]>;
<![CDATA[Nuralamsah Zulkarnaim]]>;
<![CDATA[Mahmuddin Mahmuddin]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.5914
<![CDATA[The development of Unmanned Aerial Vehicles (UAVs) has introduced new challenges in route optimization, particularly in the Drone Routing Problem (DRP), where limited battery capacity directly affects operational feasibility. This study aims to optimize DRP by considering energy constraints and charging station integration using the Improved ParthenoGenetic Algorithm (IPGA). The proposed method employs a sequence-based chromosome representation, an energy consumption model, and adaptive charging station insertion during fitness evaluation to ensure that the generated routes remain feasible under battery limitations. The performance of IPGA was evaluated using three datasets consisting of 10, 30, and 50 customers and compared with PGA, GA, and PSO based on best fitness, minimum total energy consumption, computation time, and convergence behavior. The results show that IPGA consistently achieved the highest best fitness and the lowest minimum total energy across all dataset scenarios. In the 50-customer dataset, IPGA reduced total energy consumption by 42.89%, 36.19%, and 57.32% compared with PGA, GA, and PSO, respectively. The convergence analysis also indicates that IPGA provides more stable fitness improvement, particularly in medium and large problem instances. These findings show that IPGA is effective for solving energy-constrained DRP and contributes to the development of adaptive metaheuristic optimization methods for intelligent UAV-based logistics and autonomous distribution systems.]]>
<![CDATA[Transformer-Based Intrusion Detection for Internet of Vehicles Using Multi-Strategy Feature Selection Approach ]]>
<![CDATA[Eko Arip Winanto]]>;
<![CDATA[M Riza Pahlevi B]]>;
<![CDATA[Sharipuddin Sharipuddin]]>;
<![CDATA[Dodi Sandra]]>;
<![CDATA[Febby Tri Ramadhanti]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2026.7.4.6001
<![CDATA[The rapid development of the Internet of Vehicles (IoV) has significantly increased the exposure of intelligent transportation systems to sophisticated cyber threats, particularly targeting in-vehicle communication networks such as the Controller Area Network (CAN). Conventional intrusion detection systems and traditional machine learning approaches often face limitations in capturing complex attack patterns under high-dimensional and dynamic vehicular network conditions. To address these challenges, this paper proposes a Transformer-based intrusion detection framework for the Internet of Vehicles using a multi-strategy feature selection approach. The proposed method integrates Information Gain, Principal Component Analysis, and Random Forest-based feature importance to systematically identify the most informative features from vehicular network traffic while reducing redundancy and computational overhead. A Transformer architecture with a self-attention mechanism is employed to model global dependencies and complex interactions within CAN bus data. The framework is evaluated on the CICIoV dataset using multiple data representations, including binary, decimal, and hexadecimal formats, to reflect realistic IoV communication scenarios. Experimental results demonstrate that the proposed model achieves consistently high detection performance, with accuracy and F1-score exceeding 99% under optimal feature selection configurations, while maintaining stable generalization across cross-validation folds. These findings indicate that the proposed framework provides a robust and effective solution for intelligent intrusion detection in Internet of Vehicles networks. Unlike existing approaches that rely on single feature selection strategies or raw feature inputs, the proposed framework uniquely integrates three complementary selection techniques within a unified Transformer-based architecture, addressing both feature redundancy and model scalability simultaneously. The findings of this study have significant implications for cybersecurity in intelligent transportation systems, offering a scalable and computationally efficient intrusion detection solution applicable to resource-constrained vehicular edge devices such as Electronic Control Units.]]>
