cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Hero Wintolo
Contact Email
herowintolo@stta.ac.id
Phone
-
Journal Mail Official
informatika@stta.ac.id
Editorial Address
-
Location
Kab. bantul,
Daerah istimewa yogyakarta
INDONESIA
Compiler
Published by Institut Teknologi Dirgantara Adisutjipto
ISSN : 22523839     EISSN : 25492403     DOI : 10.28989/compiler
Core Subject : Science,
Computer Science & IT
Jurnal "COMPILER" dengan ISSN Cetak : 2252-3839 dan ISSN On Line 2549-2403 adalah jurnal yang diterbitkan oleh Departement Informatika Sekolah Tinggi Teknologi Adisutjipto Yogyakarta. Jurnal ini memuat artikel yang merupakan hasil-hasil penelitian dengan bidang kajian Struktur Diskrit, Ilmu Komputasi , Algoritma dan Kompleksitas, Bahasa Pemrograman, Sistem Cerdas, Rekayasa Perangkat Lunak, Manajemen Informasi, Dasar-dasar Pengembangan Perangkat Lunak, Interaksi Manusia-Komputer, Pengembangan Berbasis Platform, Arsitektur dan Organisasi Komputer, Sistem Operasi, Dasar-dasar Sistem,Penjaminan dan Keamanan Informasi, Grafis dan Visualisasi, Komputasi Paralel dan Terdistribusi, Jaringan dan Komunikasi, Desain, Animasi dan Simulasi Pesawat Terbang. Compiler terbit setiap bulan Mei dan November.
Arjuna Subject : -
Articles 430 Documents
 39   40   41   42   43 
Multi-Label Opinion Mining Based on Random Forest with SMOTE and ADASYN Ardiansyah, Ricy; Yuliansyah, Herman; Yudhana, Anton
Compiler Vol 14, No 2 (2025): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v14i2.3185

Abstract

Multi-label classification is essential to categorize data into multiple labels simultaneously. However, data imbalance poses a challenge, where some labels have much less representation, thus reducing the model performance. This study aims to propose a candidate-based sentiment analysis model on the 2024 Jakarta Presidential and Gubernatorial Election review. The SMOTE and ADASYN oversampling methods are applied to handle class imbalance. Both oversampling methods are compared with the Random Forest machine learning method. The experimental results show that. The experimental results show that in the classification of Presidential candidates, Random Forest achieves an accuracy of 0.947 with SMOTE and 0.948 with ADASYN. For sentiment labels, the accuracy of Random Forest remains high with a result of 0.989 for both SMOTE and ADASYN. In the classification of Jakarta Gubernatorial candidates, Random Forest + SMOTE produces an accuracy of 0.975, while with ADASYN it decreases slightly to 0.973. For sentiment labels, both SMOTE and ADASYN have the highest accuracy of 0.993. The application of SMOTE and ADASYN helps to improve the distribution of the minority class without decreasing the overall accuracy, as well as improving the stability in recognizing various multi-label classes in a balanced manner.
Baseline Evaluation of Backpropagation Artificial Neural Network for Visual Image-Based Vehicle Type Classification Harman, Rika; Riadi, Imam; Fadlil, Abdul
Compiler Vol 14, No 2 (2025): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v14i2.3210

Abstract

The increasing number of vehicles in urban areas requires technology-based solutions for efficient transportation management. This study proposes a vehicle classification model using Artificial Neural Networks (ANN) with the backpropagation algorithm, based on digital image data. The model is a feedforward neural network comprising an input layer, a hidden layer with 64 sigmoid-activated neurons, and an output layer with 7 softmax-activated neurons. The dataset, sourced from Roboflow Inc., consists of 16,185 images across eight vehicle classes: Hummer, Toyota Innova, Hyundai Creta, Suzuki Swift, Audi, Mahindra Scorpio, Rolls Royce, and Tata Safari. The data is split 80:20 for training and testing. Input features include vehicle dimensions, dominant RGB color, number of axles, and license plate detection. The model is trained using gradient descent and categorical crossentropy loss. Evaluation results show 85% validation accuracy at epoch 28 and 100% test accuracy. Precision, recall, and F1-score indicate strong performance, though minor errors occur in visually similar classes. These findings demonstrate that backpropagation-based ANN is effective for vehicle classification and can be applied in systems such as automatic parking and traffic monitoring
Digital Forensic Analysis of Hybrid Scooter Motors using Smart Flow and Integrated Digital Forensics Standard Saputri, Yerly Ania; Fazal, Ahmad; Ningrat, Aditya Wahyu; Hariyadi, Dedy
Compiler Vol 14, No 2 (2025): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v14i2.3211

Abstract

This research addresses the challenges of digital forensics for connected hybrid vehicles, focusing on the Yamaha Fazzio hybrid scooter. The study highlights how limited collection methods and mobile device encryption often compromise the integrity of electronic evidence. To address these issues, a five-stage framework was developed, combining guidelines from NIST SP 800-101 Rev.1 and ISO/IEC 27037:2012. This comprehensive framework includes data collection, evidence identification, forensic acquisition, examination and analysis, and final reporting. The framework's effectiveness is boosted by Smart Flow automation on Cellebrite UFED devices, which automates the identification of Android devices, extraction of Y-Connect App databases, GPS logs in JSON, and travel route thumbnails. This automation significantly enhances the efficiency of the acquisition and analysis processes while maintaining evidence integrity. Evaluations showed successful data acquisition from a Xiaomi Mi 5s Plus with the Y-Connect App. Details from the riding_log table were extracted, providing information on travel routes, distance, average and maximum speeds, and estimated fuel consumption during a Yogyakarta - Klaten travel test scenario. These results are crucial for developing digital forensics SOPs for other connected hybrid vehicles in future research.
Topic Analysis of Indonesian Online News on the Free Nutritious Meal Program Using Non-Negative Matrix Factorization Dwijayanti, Irmma; Lahitani, Alfirna Rizqi; Habibi, Muhammad
Compiler Vol 14, No 2 (2025): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v14i2.3499

Abstract

The Free Nutritious Meal Program (MBG) represents a key policy of the Indonesian government to address malnutrition and stunting by providing nutritious meals for students. This study applies Non-Negative Matrix Factorization (NMF) for topic modeling on a long-text corpus of 5,390 digital news articles collected from seven national portals, with the aim of mapping public discourse on MBG. The optimal number of topics was determined using the coherence score, yielding nine distinct themes. Findings indicate that media coverage primarily revolves around program distribution in schools, the role of Micro, Small, and Medium Enterprises (MSMEs) and the food sector, budget allocation, political dynamics of national figures, and health-related concerns such as student poisoning cases. The results suggest that MBG is widely perceived as a strategic policy with broad implications for public policy, economic development, political debate, and social welfare. Methodologically, this research demonstrates the effectiveness of NMF in identifying latent thematic structures within long-text news corpora, offering insights into how digital media frames and interprets government initiatives.
Edge-Based IoT EndPoint Performance Optimization for Smart Mosques Rosad, Safiq; lasimin, lasimin; Irkhamudin, Faik; Pramudita, Cahyaning Yuniar; Alifah, Shofari Rizqi
Compiler Vol 14, No 2 (2025): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v14i2.3496

Abstract

This study examines the performance optimization of IoT endpoints based on edge architecture in a Smart Mosque system to improve the efficiency of managing electronic devices such as room lighting, air conditioning, and loudspeakers according to prayer schedules. The main problems faced include limited internet connectivity and the need for real-time service availability with minimal latency. The research objective is to test and implement optimization techniques such as data caching on the edge, decoupled computation, and push mechanisms using Server-Sent Events (SSE) and WebSocket to reduce latency and strengthen service availability. The methods used include prototype design with ESP32 endpoints, testing p50, p95, and p99 percentile latencies, and monitoring service availability and network resilience in baseline and optimized push mode scenarios. The test results show a significant reduction in average and tail latency, an increase in service availability approaching 99%, and system continuity during network disruptions. These findings confirm that edge computing architecture is very effective in supporting time-critical IoT systems such as Smart Mosques.
The Impact of Cloud-Based Information Systems on Organizational Performance in Education: A PIECES Framework Evaluation in Bekasi City Sumardiono Sumardiono; Rika Apriani; Sigit Setiawan; Fazril Mantovani; Reykhando Rifki Awiliyanto; Adrianus Trigunadi Santoso
Compiler Vol 15, No 1 (2026): May
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v15i1.3643

Abstract

This study aims to evaluate the impact of cloud-based information system implementation on organizational performance in the education sector, particularly in Bekasi City. The method used is descriptive-quantitative with the PIECES framework approach (Performance, Information, Economy, Control, Efficiency, Service). Data were collected through questionnaires and interviews with 10 respondents consisting of education service operators, teachers, and school operators. The results show that five PIECES dimentions—namely Performance (4.38), Information (4.22), Economy (4.20), Efficiency (4.33), and Service (4.35)—are in the very satisfactory category. However, the Control dimention only obtained a score of 2.85, which is included in the neutral category, indicating weaknesses in system security governance. This study concludes that although cloud systems have improved operational efficiency and service quality, regular information system security audits and the development of standard operating procedures (SOPs) are needed to strengthen the control dimention. The implications of this study serve as a basis for developing better information security governance in the digital transformation of education.
Behavioral Analysis Using Machine Learning Algorithms in Online Proctored Assessments; The Potential of Mobile Proctoring in E-Assessments: A Review Bartholomew Oganda Mogoi; John Kamau; Raymond Ongus
Compiler Vol 15, No 1 (2026): May
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v15i1.3680

Abstract

The rapid expansion of online learning has intensified the need for secure and reliable examination systems. While artificial intelligence (AI) and machine learning (ML) are increasingly employed to safeguard integrity through behavioral analysis, current proctoring solutions remain largely desktop-centric and poorly adapted to mobile environments, where smartphones dominate access. These limitations—spanning heterogeneous hardware, unstable lighting and connectivity, algorithmic bias, privacy concerns, and limited explainability—undermine fairness and trust in misconduct detection. This study reviews the potential of mobile-based proctoring systems that leverage ML for real-time behavioral analysis in e-assessments. A systematic search across Scopus, IEEE Xplore, SpringerLink, and Google Scholar identified 200 publications from 2015–2025. After applying inclusion and exclusion criteria, 30 peer-reviewed empirical studies were analyzed in depth. Findings reveal that ML-driven behavioral analytics enhance accuracy and fairness by detecting anomalies such as gaze aversion, facial emotion changes, and environmental inconsistencies. However, research gaps persist in mobile optimization, ethical safeguards, privacy protection, and bias mitigation. The review concludes that hybrid mobile frameworks integrating multimodal behavioral features and adaptive ML models are essential to strengthen proctoring accuracy while ensuring accessibility and privacy. Future research should prioritize context-aware and explainable AI to foster equitable and trustworthy online assessment environments.
Designing a Knowledge Management System Language Course Using SECI Model and RAD Yuni Sugiarti; Haniya Nadira; An Nisa Ramadhanti
Compiler Vol 15, No 1 (2026): May
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v15i1.2690

Abstract

This research aims to design an efficient Knowledge Management System (KMS) by applying the Rapid Application Development (RAD) method and integrating the SECI model in the context of a language course. The study emphasizes the need for a fast and iterative development process to enhance the transfer of knowledge and improve learning outcomes. The approach focuses on combining the strengths of RAD's rapid prototyping with the SECI model, which supports knowledge sharing and internalization among learners. The method used includes the design and implementation of a web-based KMS to facilitate the exchange of knowledge between instructors and students. The findings indicate that the integration of RAD and the SECI model successfully improves the effectiveness of the KMS in enhancing learning experiences. The study concludes that a well-designed KMS can significantly support knowledge transfer and improve educational quality in language courses.
Evaluating IndoBERT for Fraudulent Tweet Detection on Social Media X Imroatul Khuluqi Izzah; Imam Riadi; Abdul Fadlil
Compiler Vol 15, No 1 (2026): May
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v15i1.3979

Abstract

The spread of fraudulent content on social media X has become an important issue because perpetrators often use persuasive, urgent, and misleading language to influence users to transfer money, share personal data, or access suspicious links. This research evaluates the performance of IndoBERT for binary classification of fraud and non-fraud Indonesian-language posts on social media X using a two-stage fine-tuning design. The dataset consists of 5,235 manually labeled posts, including 2,557 fraud and 2,678 non-fraud instances. In Stage 1, four IndoBERT variants, namely indobert-base-p1, indobert-base-p2, indobert-large-p1, and indobert-large-p2, were compared using a uniform training configuration to identify the best model. The results showed that indobert-large-p1 at epoch 5 achieved the best performance, with a validation F1-score for the fraud class of 0.8898 and a test accuracy of 0.8989. In Stage 2, the selected model was re-evaluated through a controlled grid search by varying epoch, learning rate, and batch size. Although the best Stage 2 configuration improved the validation F1-score to 0.8975, it did not surpass the best Stage 1 model on the test set. These findings indicate that IndoBERT is effective for fraud detection and that a two-stage evaluation design supports more systematic model selection.
An ADDIE-Based Approach to Developing an Automatic Room Sprayer from Recycled Styrofoam Waste Rindi Nur Wulandari; Hero Wintolo; Adila Aviv Khairunisa; Parhan Parhan; Lasmadi Lasmadi; Emmanuel Endy Ismoyo; Rifky Adhi Pangestu
Compiler Vol 15, No 1 (2026): May
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/compiler.v15i1.3999

Abstract

Styrofoam waste constitutes an increasingly acute environmental problem in Indonesia due to its non-biodegradable nature and widespread accumulation across various sectors. This study presents the design, development, and evaluation of a proof-of-concept prototype of an Arduino microcontroller-based automatic room sprayer — designated Aero-Spray — that utilizes recycled styrofoam as the primary structural material, shaped into a miniature representation of an Airbus A320-200 aircraft. The system was developed following the ADDIE (Analysis, Design, Development, Implementation, and Evaluation) instructional design model, which enabled a systematic and iterative development process. The hardware architecture integrates a PIR motion sensor, servo motor, aroma spray module, LCD display, LED indicators, and a buzzer, all coordinated by an Arduino microcontroller. Comprehensive functional testing demonstrated that the PIR sensor reliably detects human movement at distances of up to 6 meters across all evaluated angles (0°, 15°, and 30°). The average system response latency from motion detection to spray activation was 3.8 seconds, while spray interval accuracy attained 98.34% of the programmed value. System power consumption was measured at 1.025 W in idle mode and 4.025 W during active spraying, yielding a total estimated daily energy consumption of 17.42 Wh to 28.16 Wh. Fragrance endurance analysis indicated operational durations of 15 to 24.5 days per 60 mL volume, depending on occupancy levels. These results collectively demonstrate that the Aero-Spray proof-of-concept prototype operates reliably under normal indoor conditions, validating the feasibility of transforming styrofoam waste into a functional, value-added, and environmentally responsible automated device.

Page 43 of 43 | Total Record : 430

 39   40   41   42   43 

Filter by Year

2012 2026


Reset
Filter By Issues
All Issue Vol 15, No 1 (2026): May Vol 14, No 2 (2025): November Vol 14, No 1 (2025): May Vol 13, No 2 (2024): November Vol 13, No 1 (2024): May Vol 12, No 2 (2023): November Vol 12, No 1 (2023): May Vol 11, No 2 (2022): November Vol 11, No 1 (2022): May Vol 10, No 2 (2021): November Vol 10, No 1 (2021): May Vol 10, No 1 (2021): Mei Vol 9, No 2 (2020): November Vol 9, No 1 (2020): Mei Vol 9, No 1 (2020): May Vol 8, No 2 (2019): November Vol 8, No 1 (2019): Mei Vol 8, No 1 (2019): May Vol 8, No 1 (2019): Mei Vol 7, No 2 (2018): November Vol 7, No 2 (2018): November Vol 7, No 1 (2018): Mei Vol 7, No 1 (2018): May Vol 7, No 1 (2018): Mei Vol 6, No 2 (2017): November Vol 6, No 2 (2017): November Vol 6, No 1 (2017): Mei Vol 6, No 1 (2017): Mei Vol 5, No 2 (2016): November Vol 5, No 2 (2016): November Vol 5, No 1 (2016): Mei Vol 5, No 1 (2016): Mei Vol 4, No 2 (2015): November Vol 4, No 2 (2015): November Vol 4, No 1 (2015): Mei Vol 4, No 1 (2015): Mei Vol 3, No 2 (2014): November Vol 3, No 2 (2014): November Vol 3, No 1 (2014): Mei Vol 3, No 1 (2014): Mei Vol 2, No 2 (2013): November Vol 2, No 2 (2013): November Vol 2, No 1 (2013): Mei Vol 2, No 1 (2013): Mei Vol 1, No 2 (2012): November Vol 1, No 2 (2012): November Vol 1, No 1 (2012): Mei Vol 1, No 1 (2012): Mei More Issue