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
962 Documents
<![CDATA[SEMAR v1.0: An AI-driven Conceptual Model and Architecture for Smart Government in Indonesia Using a Mixed-Methods Approach ]]>
<![CDATA[Arief, Assaf]]>;
<![CDATA[Muhammad, Miftah]]>;
<![CDATA[Fuad, Achmad]]>;
<![CDATA[Sensuse, Dana Indra]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.4996
<![CDATA[Building smart cities represents a national priority for Indonesia to enhance global competitiveness, with artificial intelligence (AI) driven smart government as a key enabler. However, implementation faces significant challenges including unmeasured organizational maturity, lack of service innovation, fragmented governance, and minimal citizen engagement leading to government institutions' failure in achieving smart city vision. This study aims to develop a holistic conceptual model to identify critical success factors and evaluate processes that integrate public services, fostering AI-driven smart government innovation at strategic level. This research employs mixed-methods exploratory sequential design combining qualitative techniques (Systematic Literature Review, expert interviews) with quantitative validation (citizen survey, statistical analysis). The model was constructed using Factor Analysis, Thematic Analysis, TOGAF framework, and multidimensional view with validation through triangulation, expert judgment, Focus Group Discussions, and statistical analysis. Results show a comprehensive model consisting of 6 dimensions, 17 key components, and 5-layer organizational architecture with high reliability (Cronbach's Alpha 0.709-0.866) and expert consensus (86% agreement in Fuzzy Delphi Method analysis). This framework, referred to as SEMAR v1.0 (Smart Government Nusantara), serves as a benchmark for assessing the maturity and readiness of local government institutions in Indonesia. It offers the potential to improve SPBE scores through systematic evaluation, while also providing a theoretical foundation for smart government scholarship and a practical blueprint for policy implementation.]]>
<![CDATA[Vibration Classification Of Intact And Cracked Brick Materials Using Fast Fourier Transform–Extreme Learning Machine For Structural Damage Early Detection ]]>
<![CDATA[Khoiri, Mohamad]]>;
<![CDATA[Farid, Imam Wahyudi]]>;
<![CDATA[Priambodo, Joko]]>;
<![CDATA[Rahayu, Lucky Putri]]>;
<![CDATA[Mahira, Balqis]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.4997
<![CDATA[Structural damage in buildings is often initiated by small cracks in lightweight brick elements, which, if undetected, may compromise structural safety. This study developed a vibration-based classification system using the ADXL345 accelerometer, Fast Fourier Transform (FFT), and Extreme Learning Machine (ELM) for early detection of such damage. Vibration data were collected along three axes (X, Y, and Z) with excitation frequencies ranging from 10–50 Hz. FFT analysis revealed clear distinctions between intact and cracked bricks, where cracked samples exhibited higher amplitudes and multiple resonance peaks. These frequency-domain features were then processed by ELM classifier. ELM achieved high computational efficiency and demonstrated strong predictive capability, correctly classifying 7,855 intact and 4,548 cracked samples. However, it also produced 1,879 false positives and 5,100 false negatives, resulting in an RMSE of 0.548. While the model proved more accurate in identifying intact bricks, its sensitivity to crack detection remains a challenge. Overall, FFT–ELM framework shows promising potential as a fast, non-destructive, and scalable approach for structural health monitoring, with further refinements needed to improve detection accuracy of damaged materials.]]>
<![CDATA[Efficient Evidence Reduction Technique for Mobile Forensics based on Digital Evidence Object (DEO) Model ]]>
<![CDATA[Hakim, Arif Rahman]]>;
<![CDATA[Saputri, Lisa]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.4999
<![CDATA[The Android operating system (OS) is currently the most widely used platform on smartphones, making it a critical source of digital evidence in cybercrime investigations. With its vast array of applications and features, Android OS generates and stores a significant amount of data, much of which may be relevant to criminal activities. Mobile forensics plays a crucial role in identifying and analyzing this information to produce scientifically valid evidence. However, the process of acquiring and examining data from a smartphone’s internal storage typically results in large and complex datasets that can hinder timely forensic analysis. To address this challenge, this paper proposes the implementation of the DEO Model using Python to reduce the volume of digital evidence obtained from Android-based smartphones. The DEO Model employs a structured filtering approach, narrowing the dataset to only those objects relevant to a predefined scenario. This is achieved by applying DEO parameters based on the 5W category theory (Why, When, Where, What, Who), resulting in an optimal and focused dataset. The findings demonstrate that the Python-based DEO Model significantly accelerates the mobile forensic process, and effectively reduces dataset size while both maintaining the evidence integrity and the scenario relevance. The model achieves a very low False Positive Rate (FPR) of 0,00072, indicating a minimal risk of mismatches during the object reduction process. Therefore, the findings confirm the validity and accuracy of the digital evidence obtained. This research highlights the potential of the Python-based DEO Model to enhance the efficiency of forensic investigations on Android smartphones.]]>
<![CDATA[Analysis of Polyglot Obfuscation Techniques against ModSecurity in Preventing Cross-Site Scripting (XSS) and SQL Injection Attacks with Experimental Method ]]>
<![CDATA[Nelmiawati, Nelmiawati]]>;
<![CDATA[Dealova, Kessy]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5000
<![CDATA[Internet use has increased every year, as shown by the percentage of internet users in Indonesia reaching 79.50% in 2024. However, security is something that cannot be ignored, especially with the growing number of Cross-Site Scripting (XSS) and SQL Injection Attacks in web platforms. According to OWASP Top 10 report, these two attacks were listed in 2017 and appeared again in the 2021 version, showing that they are still relevant today. In fact, in June 2024, XSS and SQL Injection vulnerabilities were found in a company, PT. XYZ. One way to mitigate these attacks is by using a Web Application Firewall (WAF) such as ModSecurity, which can protect websites from exploitation. However, previous research found that older versions of ModSecurity had weaknesses that could be bypassed with simple obfuscation techniques. This study aims to analyze the effectiveness of the built-in rules in ModSecurity Core Rule Set (CRS) version 4.7 in handling XSS and SQL Injection payloads with polyglot obfuscation, a method that uses complex character encoding to avoid WAF detection. The research was conducted using an experimental method. This study contributes to improve WAF security by testing against modern obfuscation-based attacks, so that security does not rely solely on the default WAF configuration. The results show that all payloads were detected and blocked by ModSecurity with an HTTP 403 response, proving that the CRS 4.7 built-in rules can effectively protect against XSS and SQL Injection threats.]]>
<![CDATA[A Comparative Analysis of Color Channel-Based Feature Extraction using Machine Learning versus Deep Learning for Food Recognition ]]>
<![CDATA[Sari, Yuita Arum]]>;
<![CDATA[Nugraha, Dwi Cahya Astria]]>;
<![CDATA[Adinugroho, Sigit]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5001
<![CDATA[Automated Dietary Assessment Accurate food recognition is a big challenge in computer vision which is critical for developing Automated Dietary assessment and health monitoring systems. The key question it answered was whether traditional machine learning with feature engineering by hand can beat modern deep learning approaches? In this Context, this study serves as a comparative analysis of these two paradigms. The baseline method worked by extracting texture (LBP,GLCM) and color information from different channels of five colors spaces (RGB, HSV, LAB, YUV,YCbCr) followed by feeding these features into multiple classifiers such as Nearest Neighbor(NN), Decision Tree and Naïve Bayes. These were then compared to deep learning models (MobileNet_v2, ResNet18, ResNet50, EfficientNet_B0). The best traditional one can reach an accuracy of 93.33%, using texture features extracted from the UV channel and classified with a NN. Nevertheless, the deep learning models consistently presented higher performance and MobileNet_v2 reached up to 94.9% accuracy without requiring manual feature selection. In this paper, we show that end-to-end deep learning models are more powerful and error robust for food recognition. These results highlight their promise for constructing more effective and scalable real-world applications with less need for intricate, domain-specific feature engineering.]]>
<![CDATA[A Hybrid Approach for Recommender Systems Based on Alternating Least Squares and CatBoost ]]>
<![CDATA[Yusfida A'la, Fiddin]]>;
<![CDATA[Hartatik, Hartatik]]>;
<![CDATA[Riasti, Berliana Kusuma]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5002
<![CDATA[This study aims to improve the accuracy of movie rating predictions by applying and combining collaborative filtering and machine learning techniques in a hybrid recommender system. The research utilizes the MovieLens dataset to implement two distinct approaches: the Alternating Least Squares (ALS) matrix factorization model and the CatBoost gradient boosting model. The ALS model is trained to capture latent user–item interactions, while CatBoost leverages nonlinear relationships using user and item features. A simple hybrid strategy averages the predictions from both models to evaluate potential performance gains. Experimental results show that the hybrid approach achieves lower error metrics compared to either model individually, with Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) values of 0.828 and 0.666, respectively. This demonstrates that combining latent factor models with tree-based learning can effectively reduce prediction errors by exploiting complementary strengths. The novelty of this research lies in its efficient yet effective hybridization strategy that improves recommendation quality without complex ensembling techniques. The findings suggest that even lightweight model fusion can significantly enhance predictive accuracy in recommender systems and may be adapted for other domains where combining linear and nonlinear modeling is beneficial. This research contributes to the field of Informatics and Computer Science by demonstrating that a lightweight hybridization of latent factor models and tree-based learning can significantly improve recommender system accuracy while offering practical implications for real-world digital applications.]]>
<![CDATA[Improving Direct Image Regression for Blood Cell Enumeration with a Fine-Tuned Backbone ]]>
<![CDATA[Adinugroho, Sigit]]>;
<![CDATA[Sari, Yuita Arum]]>;
<![CDATA[Utaminingrum, Fitri]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5004
<![CDATA[Complete blood count (CBC) examination provides an important insight for diagnosis or disease treatment. Currently, CBC examination requires complex and expensive devices that limit their deployment in remote area. The development of computer vision based method offers simplification to the process. However, its implementation is limited to the availability of large size labelled dataset. This research aims to develop a direct image regressor that is able to regress directly from image. There are two stages in estimation process. First, the backbone is trained using large dataset available for blood cell classification problem. Then the trained backbone is plugged into the final model by adding a fully connected neural network that acts as regressor. The whole model is then trained using limited whole blood cell count dataset. The evaluation process shows that training the backbone using large size related dataset improve the performance by 50%. This study can be used to create a low-cost blood component evaluation tool, particularly in rural areas where access to advanced laboratory equipment is limited.]]>
<![CDATA[Comparison of Accuracy and Computation Time for Predicting Earthquake Magnitude in Java Island ]]>
<![CDATA[Yuniarto, Abdul Hakim Prima]]>;
<![CDATA[Hariguna, Taqwa]]>;
<![CDATA[Nawangnugraeni, Devi Astri]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5044
<![CDATA[Java Island has numerous active faults, making earthquake magnitude prediction a crucial component of disaster mitigation efforts. This study conducted a rigorous comparative analysis of four machine learning algorithms—Random Forest, Neural Network, Linear Regression, and Support Vector Machine—to determine their effectiveness in this specific task. The methodology employed involved systematic hyperparameter optimization for each model to ensure a fair and robust evaluation, with performance measured by Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and training time. The results showed that all three nonlinear models significantly outperformed Linear Regression. Random Forest achieved the highest accuracy (RMSE 0.5445), but Support Vector Machine and Neural Network demonstrated very competitive and nearly equal performance. The study concluded that while Random Forest has a slight advantage, several state-of-the-art models are highly capable of addressing this problem after appropriate optimization. This underscores the critical role of methodical tuning and implies that model selection in practical applications depends on a trade-off between modest improvements in accuracy and computational efficiency.]]>
<![CDATA[SN Systemic Integration of Artificial Intelligence in Indonesian Television Using Soft Systems Methodology ]]>
<![CDATA[Setyobudi, Ciptono]]>;
<![CDATA[Damayanti, Ratih]]>;
<![CDATA[Setiawan IS, Teguh]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5047
<![CDATA[The television industry faces significant challenges due to digital disruption, particularly the increasingly widespread penetration of artificial intelligence (AI) technology. AI has the potential to optimize production, distribution, and audience preference analysis, but its implementation faces the complexities of unstructured social systems. This study aims to explore the systemic application of AI using a Soft Systems Methodology (SSM) approach to identify, analyze, and optimize its use in the television industry. The research method used was qualitative with a case study design at a national television station. The SSM process was carried out through seven stages, starting from exploring the problem situation, compiling rich pictures, analyzing CATWOE, and formulating and evaluating corrective actions. Data were collected through literature review, participant observation, and internal document analysis. The results show that SSM is effective in identifying strategic areas for AI optimization, particularly in audience segmentation, content automation, and broadcast management. The resulting framework is flexible, participatory, and responsive to the social dynamics of media organizations. The impact of this research is a contribution to the development of media information systems and technology, as well as expanding the scope of soft methodologies in responding to the challenges of digital transformation in an adaptive and sustainable manner.]]>
<![CDATA[Evaluating Classification Models for Predicting Product Success in Indonesian E-Commerce ]]>
<![CDATA[Aulya, Fiola Utri]]>;
<![CDATA[Kusnawi, Kusnawi]]>
<![CDATA[ Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025 ]]>
Publisher : <![CDATA[Informatika, Universitas Jenderal Soedirman ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52436/1.jutif.2025.6.4.5071
<![CDATA[The intense competition within the Indonesian e-commerce landscape presents a significant challenge for sellers in forecasting product performance. This study offers a unique contribution by systematically comparing seven machine learning classification algorithms to predict product success across Indonesia's three largest platforms: Shopee, Tokopedia, and Lazada. The primary objective is to identify the most effective algorithm for predicting whether a product's sales will surpass the market median. The methodology involved aggregating and preprocessing a dataset of 3,673 product listings. Product success was defined as a binary variable based on sales volume exceeding the dataset's median. Seven models, including Logistic Regression, KNN, SVM, and tree-based ensembles like Random Forest, XGBoost, and LightGBM, were trained and optimized using a 5-fold cross-validated GridSearchCV. Evaluation was based on accuracy, ROC AUC, and F1-score. The results demonstrate a clear performance hierarchy, with tree-based ensemble models achieving superior results. Random Forest emerged as the premier model, attaining an accuracy of 83.2% and an AUC of 0.907. A subsequent feature importance analysis revealed that shop_followers and price were the most significant predictors of success. This finding has crucial practical implications, particularly for Micro, Small, and Medium Enterprises (MSMEs), by providing a data-driven framework for decision-making. The model enables them to focus resources on actionable strategies—building seller reputation and optimizing pricing—to enhance their competitiveness effectively.]]>
