cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Alam Rahmatulloh
Contact Email
alam@unsil.ac.id
Phone
+6285223519009
Journal Mail Official
innovatics@unsil.ac.id
Editorial Address
Program Studi Informatika Fakultas Teknik Universitas Siliwangi Jl. Siliwangi No. 24 Tasikmalaya, Jawa Barat
Location
Kota tasikmalaya,
Jawa barat
INDONESIA
Innovation in Research of Informatics (INNOVATICS)
Published by Universitas Siliwangi
ISSN : -     EISSN : 26568993     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering
Innovation in Research of Informatics (Innovatics) merupakan Jurnal Informatika yang bertujuan untuk mengembangkan penelitian di bidang: Machine Learning Computer Vision Internet of Things Information System and Technology Natural Language Processing Image Processing Network Security Geographic Information System Knowledge based Computer Graphic Cyber Security IT Governance Data Mining Game Development Digital Forensic Business Intelligence Pattern Recognization Virtual & Augmented Reality Virtualization Enterprise Application Self-Adaptive Systems Human Computer Interaction Cloud Computing Mobile Application Innovatics adalah jurnal peer-review yang ditulis dalam bahasa Indonesia yang diterbitkan dua kali dalam setahun mulai dari Vol. 1 No.1 Maret 2019 (Maret, dan September) dengan proses peninjauan menggunakan double-blind review.
Arjuna Subject : Ilmu Komputer (semua kategori) - Sistem Informasi
Articles 93 Documents
 6   7   8   9   10 
An Integrated Convolutional Neural Networks and Light Gradient Boosting Approach for Flood Classification Using Sentinel-1 SAR Satellite Imagery Anshori, Siddiq Ahmad; Hadiana, Asep Id; Kasyidi, Fatan
Innovation in Research of Informatics (Innovatics) Vol 7, No 1 (2025): March 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i1.13600

Abstract

od classification plays a crucial role in disaster mitigation, particularly in areas frequently affected by floods. This study proposes a novel model combining Convolutional Neural Networks (CNN) using ResNet-50 and Light Gradient Boosting Machine (LightGBM) for classifying flood and non-flood areas using Sentinel-1 SAR imagery. The dataset used consists of 21,016 images, evenly distributed between flood and non-flood classes, and processed through resizing, normalization, denoising, and augmentation. Feature extraction was conducted using the ResNet-50 architecture, which captured spatial and textural patterns efficiently, followed by LightGBM for classification. The proposed model achieved a high accuracy of 96%, with Precision, Recall, and F1-scores exceeding 95% for both classes. The evaluation metrics, including Precision-Recall Curve with an AUC of 0.9852 and a Confusion Matrix, confirmed the model's robustness and balance in classifying both categories. Additionally, comparisons with previous research, such as SAR-FloodNet, demonstrated the superiority of the proposed approach, achieving a 2% improvement in accuracy. Despite these results, limitations such as the exclusive use of Sentinel-1 data and the lack of validation across diverse environmental conditions remain. Future research should explore integrating multispectral Sentinel-2 data and testing on broader datasets to enhance scalability and reliability. The findings underscore the model's potential for real-world applications in flood monitoring and disaster management systems.
Prediction of Alternative Solar Energy Utilization in Internet of Things Based Systems Using Random Forest Algorithm Natsir, Fauzan; Abdurahman, Abdurahman; Sihombing, Redo Abeputra
Innovation in Research of Informatics (Innovatics) Vol 7, No 1 (2025): March 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i1.13096

Abstract

The continuous use of fossil energy is depleting available energy sources, necessitating the adoption of alternative or renewable energy sources, such as solar energy. With rapid technological developments, particularly in Internet of Things (IoT) applications such as ThingSpeak, new monitoring solutions are becoming available. ThingSpeak can be used to monitor and control device outputs efficiently. Controlling the work of the tool is one of the efforts to save energy. In this study utilizing the ThinkSpeak application with the support of the ESP 8266 component to send information obtained through the sensor to be displayed on the monitor screen and set the output on or off. So that electricity is used as needed. The results of this study show that the output of sensor readings that appear from the serial monitor via the Arduino application is almost the same as the display on ThinkSpeak. So that this tool can support energy saving both in terms of solar energy utilization, the tool work control system for utilization according to needs and can also monitor the condition of the battery whether it is still in good or bad condition
Integration of SMOTE and Ensemble Models for Predicting Airline Passenger Satisfaction Laksono, Bagus; Kurniawati, Ika; Sriwiyanta, Aditya Budi; Zaenudin, Zen Zen; Ramadha, Johan Afrian; Alfian, Desri
Innovation in Research of Informatics (Innovatics) Vol 7, No 1 (2025): March 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i1.14001

Abstract

The high public interest in air transportation has become a polemic for airline companies that are competing to maintain their existence by continuously improving their services. The passenger satisfaction survey data collected has several problems such as unbalanced data, missing values, noise, difficulty finding significant patterns and biased data. Imbalanced class causes the classification results to lean more towards the majority class, this can reduce the performance of the prediction model. SMOTE is one of the over-sampling methods to balance the dataset by increasing the number of samples in the minority class based on k-nearest neighbors to approach the same class. Boosting is a machine learning strategy that combines many very fragile and poor prediction rules to produce very accurate prediction rules. In this study, we conducted a model experiment by integrating the SMOTE and AdaBoost ensembles with the classification algorithm to obtain the best performance metrics. The results showed that the performance of integrating the DT + SMOTE and DT + SMOTE + AdaBoost models produced an accuracy of 91.88%, this performance is superior to the traditional DT model. Significant performance improvements also occur in the integration of NB+SMOTE+AdaBoost and NB+AdaBoost, which is an increase of around 5% compared to NB. However, the application of SMOTE to NB decreases accuracy because SMOTE produces synthetic samples that can disrupt the independence assumption of NB. The results of this study demonstrate the superiority of our proposed method, a robust ensemble learning compared to traditional machine learning classifiers. Both techniques are very efficient in improving classification capabilities, especially in cases of complex and imbalanced data. AdaBoost, Customer satisfaction prediction, Data mining, Ensemble learning, Imbalanced data, SMOTE. 
Melanoma Skin Cancer Classification Using EfficientNetB7 for Deep Feature Extraction and Ensemble Learning Approach Darmawan, Aditya Yoga; Dullah, Ahmad Ubai; Qohar, Bagus Al; Unjung, Jumanto; Muslim, Much Aziz
Innovation in Research of Informatics (Innovatics) Vol 7, No 1 (2025): March 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i1.12764

Abstract

Cancer is one of the deadliest diseases in the world. cancer is caused by the presence of cancer cells due to abnormal conditions during the cell turnover process. One of the dangerous types of cancer is melanoma skin cancer, this cancer attacks the outer skin of humans because skin cells are prone to damage. However, diagnosis for this disease is mostly done manually while there are previous studies that use deep learning approaches with the accuracy that can be improved. The purpose of this study is to find an effective and efficient method for melanoma cancer recognition so that it can be treated more quickly. We propose several methods that we have compared to be able to classify melanoma skin cancer with EfficientNetB7 Feature Extractor and Ensemble Learning. The results of this research model get the highest accuracy of 91.2%. When EfficientNetB7 together with ensemble learning. This research model has better and efficient results when compared to previous research.
Automated Identification of Oil Palm’s 17th Leaf Using YOLOv12 and Spatial Positioning Rahmawan, Jihad; Yuliansyah, Herman; Yudhana, Anton; Irfan, Syahid Al
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.15766

Abstract

This study proposes an artificial intelligence–based approach for automatic identification of the 17th leaf in oil-palm trees (Elaeis guineensis), which serves as a key physiological indicator for nutrient monitoring. The method integrates YOLOv12 object detection with a spatial-positioning algorithm that estimates leaf order through vertical sorting of detected fronds. A total of 1,250 annotated field images were collected from farmer-recorded videos to train and evaluate the system. The proposed model achieved a mean average precision (mAP@0.5) of 92.4% and an average positional error of 10.6 pixels in locating the 17th leaf. Compared with manual identification that requires 3–5 minutes per tree, the automated system performs the entire process in under 15 seconds, providing over 95% time efficiency improvement. This work demonstrates a novel fusion of real-time deep-learning detection and spatial reasoning for nutrient-focused precision agriculture and establishes a practical foundation for scalable, automated leaf indexing in plantation management.
Comparison of Efficiency and Security of AES, Blowfish, and ChaCha20 Cryptographic Algorithms on Image and Document Files Bintang Timur, Muhammad Bagus; Royansyah, Royansyah; Kusumaningsih, Dewi
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.15748

Abstract

This study presents a comparative analysis of the Advanced Encryption Standard (AES), Blowfish, and ChaCha20 cryptographic algorithms in terms of their efficiency and security when applied to image and document files. The growing need for secure data transmission and storage has highlighted the importance of selecting suitable encryption algorithms based on specific file types and operational contexts. In this research, various sample files in image (JPG, PNG) and document (PDF, DOCX) formats were encrypted using the three algorithms. Performance was measured based on encryption and decryption time, CPU and memory usage, and file size changes after encryption. Security evaluation focused on resistance to brute-force attacks, key length strength, and algorithmic robustness. The experimental results indicate that ChaCha20 demonstrated superior performance in terms of processing speed and resource efficiency, especially on low-power systems. AES, while slightly slower, offered a high level of security and is widely trusted for sensitive document encryption. Blowfish, although fast, showed limitations in handling larger files and is less favored due to its shorter block size. This study concludes that no single algorithm is universally optimal; instead, the choice should depend on the specific use case, balancing efficiency and security requirements. These findings provide useful insights for practitioners and researchers in selecting appropriate cryptographic methods for protecting digital assets.
Forest Biomass Estimation through the Integration of UAV Imagery and Vegetation Indices: Toward Accurate and Efficient Monitoring Fadilah, Vira Hasna; Hadiana, Asep Id; Komarudin, Agus
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.14931

Abstract

Forest biomass estimation method using drone imagery and vegetation index, focusing on the effectiveness and efficiency of the approach. Using high-resolution drone imagery, this study analyzes vegetation structure and density, and supports the development of a more accurate biomass estimation model compared to traditional methods. Drone imagery has the advantage of collecting data quickly and in real time, especially in areas that are difficult to access manually. Vegetation indices, such as NDVI, are used to assess vegetation health and density, which are closely related to biomass estimation. The combination of drone imagery and vegetation indices can produce more detailed data, support 3D vegetation modeling, and help estimate biomass volume over time. This study is expected to produce data and biomass estimation models that support sustainable forest management as well as technical recommendations for the use of drones for vegetation monitoring. The findings of this study show that the proposed method produces an estimation accuracy of 85.2% based on field validation data calculated using simple linear regression. The findings of this study are expected to make a significant contribution to the development of drone-based technology for efficient and environmentally friendly natural resource management.
Liquefied Petroleum Gas (LPG) Leak Detection Mitigation System with MQ-6 Sensor based on the Internet of Things (IoT) Novitaningrum, Dian; Handayani, Yuni; Hidayat, Taufik
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.16705

Abstract

The community is beginning to shift from the use of petroleum fuel to Liquefied Petroleum Gas (LPG). In 2023, the Kendal Regency Statistics Agency recorded 53 cases of fire. One of the factors contributing to these fires was gas cylinder leaks, which require preventive measures, education, and mitigation efforts for the proper use of LPG. This research was conducted by designing an LPG gas leak detection system based on the Internet of Things (IoT) using an MQ-6 sensor to notify users of emergencies. The systems aims to notify users via the Blynk application to prevent gas leaks. The research method includes designing the device by assembling and testing components. Additionally, software was developed to connect the sensor to the notification application using Blynk. The system can detect LPG gas leaks within a range of 1-16 cm. A safe threshold is defined as gas levels < 40 ppm, while levels >45 ppm indicate a hazardous status. The conclusions from this research shows that the average gas concentration when the green LED is on 33 ppm with a detection time of 0 seconds, the yellow LED at 40.6 ppm with a detection time of 11.6 seconds, and the red LED at 50 ppm with a detection time of 25.3 seconds, accompanied by a buzzer sounding as a warning of a gas leak in the LPG cylinder. Further research focused on improving the accuracy of the system connected to users WhatsApp accounts.
Data Augmentation Strategies on Spectrogram Features for Infant Cry Classification Using Convolutional Neural Networks Alam, Alam; Setyoningrum, Nuk Ghurroh; Maududy, Robby; Damayanti, Dea Dewi; Rahmawati, Hilmi
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.16823

Abstract

Infant cry classification is an important task to support parents and healthcare professionals in understanding infants’ needs, yet the challenge of limited and imbalanced datasets often reduces model accuracy and generalization. This study proposes the application of diverse audio data augmentation strategies including time stretching, time shifting, pitch scaling, and polarity inversion combined with spectrogram representation to enhance Convolutional Neural Network (CNN) performance in classifying infant cries. The dataset from the Donate-a-Cry Corpus was expanded from 457 to 6,855 samples through augmentation, improving class balance and variability. Experimental results show that CNN accuracy increased from 85% before augmentation to 99.85% after augmentation, with precision, recall, and F1-score reaching near-perfect values across all categories. The confusion matrix further confirms robust classification with minimal misclassifications. These findings demonstrate that data augmentation is crucial to overcoming dataset limitations, enriching acoustic feature diversity, and reducing model bias, while offering practical implications for the development of accurate, reliable, and real-world applicable infant cry detection systems.
Design of Prototype Early Warning System for Ship Gas Leakage and Fire Using MQ-6 Sensor and Arduino Uno Microcontroller Chandra, Yudi Irawan; Sjafrina, Fitri; Irawati, Diyah Ruri; Riastuti, Marti
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.16382

Abstract

Maritime transportation is vital for global trade but faces risks from gas leaks and fires, which can endanger ships and crews. Many small and medium-sized vessels lack effective systems to detect hazardous gases early, raising the chance of accidents. This study focuses on developing a prototype early warning system to detect gas leaks and fires on ships using an MQ-6 sensor and an Arduino Uno microcontroller. The goal is to provide a low-cost, real-time monitoring solution to improve maritime safety. The research follows a hardware prototyping and software development process using a waterfall approach, including requirement analysis, system design, implementation, testing, and deployment. The MQ-6 sensor identifies flammable gases like LPG and butane, while the Arduino Uno processes the sensor data and triggers alarms when gas levels become dangerous. The prototype was tested in a simulated ship environment to assess its sensitivity, response time, and reliability. Test results show that the system detects gas concentrations above safe limits within 3 to 5 seconds, activating visual and audio alarms quickly. The device also operates continuously with low power consumption, making it suitable for long-term use on ships. The study shows that this prototype can serve as an affordable and reliable additional safety device for maritime settings. This research supports efforts to improve ship safety systems by offering a practical way to reduce risks from gas leaks and fires. The system’s simplicity and effectiveness make it a promising option for enhancing onboard hazard detection.

Page 9 of 10 | Total Record : 93

 6   7   8   9   10 

Filter by Year

2019 2025


Reset
Filter By Issues
All Issue Vol 7, No 2 (2025): September 2025 Vol 7, No 1 (2025): March 2025 Vol 6, No 2 (2024): September 2024 Vol 6, No 1 (2024): March 2024 Vol 5, No 2 (2023): September 2023 Vol 5, No 1 (2023): Maret 2023 Vol 4, No 2 (2022): September 2022 Vol 4, No 1 (2022): Maret 2022 Vol 3, No 2 (2021): September 2021 Vol 3, No 1 (2021): Maret 2021 Vol 2, No 2 (2020): September 2020 Vol 2, No 1 (2020): Maret 2020 Vol 1, No 2 (2019): September 2019 Vol 1, No 1 (2019): Maret 2019 More Issue