cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Muhammad Nur Faiz
Contact Email
faiz@pnc.ac.id
Phone
+6282324039994
Journal Mail Official
jinita.ejournal@pnc.ac.id
Editorial Address
Department of Informatics Engineering Politeknik Negeri Cilacap Jln. Dr.Soetomo No.01 Sidakaya, Cilacap, Indonesia
Location
Kab. cilacap,
Jawa tengah
INDONESIA
Journal of Innovation Information Technology and Application (JINITA)
Published by Politeknik Negeri Cilacap
ISSN : 27160858     EISSN : 27159248     DOI : https://doi.org/10.35970/jinita.v2i01.119
Core Subject : Science, Engineering,
Computer Science & IT Decision Sciences, Operations Research & Management Engineering
Software Engineering, Mobile Technology and Applications, Robotics, Database System, Information Engineering, Interactive Multimedia, Computer Networking, Information System, Computer Architecture, Embedded System, Computer Security, Digital Forensic Human-Computer Interaction, Virtual/Augmented Reality, Intelligent System, IT Governance, Computer Vision, Distributed Computing System, Mobile Processing, Next Network Generation, Natural Language Processing, Business Process, Cognitive Systems, Networking Technology, and Pattern Recognition
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 191 Documents
 16   17   18   19   20 
Comparison of Multiple Linear Regression, Random Forest, and Decision Tree Methods for House Price Prediction in West Nusa Tenggara Mindi Richia Putri; Asmaul Husna RS; Ahmad Fatoni Dwi Putra; Muhammad Ari Rifqi; Syaidatussalihah
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3010

Abstract

This study aims to develop an accurate housing price prediction model for West Nusa Tenggara (NTB) by comparing three machine learning approaches: Multiple Linear Regression (MLR), Random Forest (RF), and Decision Tree (DT). The research adopts a quantitative experimental design consisting of data collection through web scraping, preprocessing, model training, and performance evaluation. Secondary data were gathered from Lamudi, Rumah123, and 99.com, containing 641 housing records with key features such as land area, building area, number of bedrooms, number of bathrooms, and price. Data preprocessing involved handling missing values, normalization, and data augmentation to improve model robustness. Model performance was assessed using Root Mean Square Error (RMSE) and the coefficient of determination (R²). The results show that non-linear models outperform linear ones, particularly after data augmentation. Without augmentation, Random Forest achieved the best performance (R² = 0.890, RMSE = 5.6×10⁹), while with augmentation, the Decision Tree model achieved the highest accuracy (R² = 0.999, RMSE = 1.1×10⁹). In contrast, MLR exhibited poor performance and sensitivity to data complexity. These findings conclude that tree-based models, particularly Decision Tree and Random Forest, provide superior prediction accuracy and adaptability for non-linear housing market data. The study contributes to the development of data-driven tools that can support policymakers, developers, and consumers in achieving more transparent and informed decision-making in the regional housing market.
Integrating Real-Time Facial Ethnicity Classification with a Godot Game Client for Personalized Cultural Interactive Exhibition Trisna Gelar; Muhammad Rizqi Sholahuddin; Aprianti Nanda Sari; Ais Laksana; Satryo Haryo; Rafli Fadhilah; Gianluigi Julian; Daffa Muzhaffar
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3021

Abstract

Addressing the recognized limitation of non-adaptive "one-size-fits-all" experiences in cultural and technical exhibitions, a novel, real-time system is introduced that integrates facial ethnicity classification with the Godot game engine to provide a personalized cultural interactive learning. The core objective was to design and quantitatively assess a performant artifact capable of dynamically classifying users into one of three major Indonesian regional categories (Barat, Tengah, or Timur) and instantly loading the appropriate personalized cultural gaming scenario. The study presents three significant contributions. First, an advanced machine learning model for Indonesian ethnicity categorization was created by utilizing an optimized feature combination of Gray Level Co-occurrence Matrix (GLCM) and Histogram of Oriented Gradients (HOG). The optimal feature set (GLCM+HOG) attained a high Cross-Validation Accuracy 95.82%±1.11%, significantly advancing the technical performance baseline for this domain. Second, a robust, real-time pipeline was successfully demonstrated, connecting the Python-based machine learning backend with the Godot gaming client via an API for immediate content customization. While the backend demonstrates high theoretical efficiency, achieving a throughput of approximately 46 FPS 21.51 ms classification latency), the final integrated system operates at a stable 12.0 FPS interface rate. This performance disparity highlights that integration overhead and feature dimensionality are the primary bottlenecks affecting the responsiveness of the on-demand classification event. Future endeavors will concentrate on enhancing real-time performance by lowering classification latency via feature dimensionality reduction (e.g., Local Binary Pattern variants) and conducting qualitative assessments to gauge user involvement and the transfer of cultural information.
Dual-Domain Transformer-Based Video Deepfake Detection Kesya Wangsa; Angelina Pramana Thenata
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3023

Abstract

Deepfake-enabled fraud caused nearly US$900 million in global losses during 2025, yet existing detection methods struggle with high-quality synthetic videos. Traditional CNN-based approaches fail to capture subtle global manipulations, while transformer-based models focus predominantly on spatial features, missing frequency-domain artifacts where manipulation traces are most evident. This research developed a dual-domain deepfake detection system combining spatial RGB analysis with Discrete Cosine Transform (DCT) frequency features, integrated into  Data-Efficient Image Transformer (DeiT-Small) architecture. The proposed DeiT-DCT model processes four-channel inputs (RGB + DCT) enabling simultaneous learning of spatial textures and frequency anomalies. Training employed the AdamW optimizer with Cosine Annealing Warm Restart, Mixup augmentation with SoftTargetCrossEntropy loss, and domain-balanced sampling via WeightedRandomSampler. Evaluated on a combined dataset of five benchmarks (Celeb-DF v2, DeeperForensics-1.0, FaceForensics++, Korean Deepfake and Indonesia, totaling 3000 videos), the model achieved 92.54% accuracy, 91.69% precision, 93.33% recall, and 92.50% F1-score on the test set. These findings demonstrate that integrating dual-domain representation with data-efficient transformer architectures produces a robust deepfake detection system deployable in real-world scenarios where manipulation techniques continuously evolve.
EMOGRAM-CNN: A Gram-Correlation Enhanced Multi-Kernel Convolutional Network for Text Emotion Recognition Marselina Endah Hiswati; Ema Utami; Kusrini Kusrini; Arief Setyanto
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3043

Abstract

Deep neural architectures have demonstrated substantial capability for handling temporal and sequential data; however, most recurrent-based models, such as LSTM, BiLSTM, GRU, and BiGRU, remain computationally expensive and prone to overfitting. This study proposes and evaluates the EMOGRAM-CNN model, a convolutional neural architecture enhanced with Gram-matrix feature correlation, to improve feature representation in temporal classification tasks. Model performance was compared with conventional CNNs and recurrent architectures on a balanced six-class dataset comprising 17,967 samples. Experimental results show that EMOGRAM-CNN achieved the highest classification accuracy of 94.48%, outperforming CNN (94.00%), GRU (92.00%), BiGRU (91.00%), BiLSTM (91.00%), and LSTM (90.00%). The model converged faster, with smoother loss behavior and lower validation error, indicating superior stability and generalization. The Gram-based correlation layer effectively preserved second-order dependencies across feature maps, enabling the network to capture both local and global temporal relationships without recurrent connections. These findings confirm that EMOGRAM-CNN offers a robust, computationally efficient alternative to recurrent deep networks for sequence classification.
Impact of Stopword Variation on Qur'anic Text Classification using Support Vector Machine and Backpropagation Afit Ajis Solihin; Fandy Setyo Utomo; Azhari Shouni Barkah
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3069

Abstract

This study aims to analyze the impact of varying stopword sets on the performance of Qur'anic text classification models in Indonesian translations, using two machine learning algorithms: Support Vector Machine (SVM) and Backpropagation Neural Network (BPNN). The research involved six stopword variants: Sastrawi, Damian Doyle, Fadillah Z. Tala, Natural Language Toolkit (NLTK) Indonesian, Yudi Wibisono, and a combination of all these lists. The preprocessing steps included cleaning, case folding, tokenization, stopword removal, and stemming, followed by TF-IDF (Term Frequency-Inverse Document Frequency) text representation. Feature selection was performed using the Chi-Square method to select the top 1,000 features. The evaluation results showed that SVM consistently outperformed BPNN across all metrics, including accuracy, precision, recall, and F1-score. The Sastrawi stopword variant delivered the best performance with an F1-score of 0.6697, followed by Fadillah Z. Tala and Damian Doyle. In contrast, BPNN showed lower performance, with the highest F1-score of 0.4607 achieved using the NLTK stopword variant. These findings highlight that selecting relevant, contextually appropriate stopwords is critical to classification Effectiveness. SVMs proved more reliable at handling high-dimensional text data while preserving the semantic meaning of Qur'anic verses.
Performance Analysis of IoT-Based pH and EC Control System for Smart Indoor Hydroponics Pagoda Mustard Cultivation Bagus Muhammad Akbar; Shoffan Saifullah; Arif Umami
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3080

Abstract

Indoor hydroponic systems increasingly utilize IoT technologies for automated nutrient management, yet empirical performance data from extended operational trials remain scarce. This study quantifies control system performance and biological validation for an IoT-based pH-EC control implementation during 35-day continuous operation. The system employed ESP32 microcontroller with safety-based threshold control (pH <8.2, TDS >900 ppm) optimized for reliability rather than precision in a 30L Deep Flow Technique system cultivating 14 pagoda mustard plants. Analysis of 12,219 datapoints revealed TDS control achieved 99.3% time-in-safe-range (MAE 50.06 ppm, settling time 671.6 min) while pH control maintained MAE 0.74 units (settling time 1266.9 min) without exceeding critical thresholds. Plant outcomes validated control effectiveness: final height 20.59 ± 2.03 cm, leaf count 35.50 ± 11.83, with consistent SPAD values >40 indicating adequate nitrogen nutrition despite sub-optimal TDS levels (867 ppm vs. optimal 1050-1400 ppm). Results demonstrate that simplified rule-based control achieves satisfactory biological outcomes when safety thresholds prevent metabolic stress, contributing validated operational benchmarks for resource-constrained hydroponic applications.
E-SAFE: EfficientNet with Squeeze-and-Attention Feature Enhancement for Deepfake Detection Rianto Rianto; Neng Ika Kurniati; Aradea Aradea; Pandu Pangestu; Irsalina Yumna
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3084

Abstract

Deepfakes have become a serious threat to digital security, individual privacy, and the spread of disinformation globally. The main challenge in detecting manipulated media lies in balancing accuracy, model complexity, and generalization across varying levels of compression. This study proposes the E-SAFE (EfficientNet with Squeeze-and-Attention Feature Enhancement) model, a deepfake detection model integrating the EfficientNet-B0 architecture with the Squeeze-and-Excitation (SE) attention mechanism. This study adopted FaceForensics++ as a benchmark dataset for evaluating deepfake detection. The model was trained with the Adam optimizer and evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and Grad-CAM-based interpretability metrics. Experimental results indicated that E-SAFE attained 95% accuracy, 94% precision, 93% recall, 93% F1-score, and 98% ROC-AUC. The results surpassed the baseline EfficientNet-B0 while maintaining high computational efficiency. These results suggest that integrating the Squeeze-and-Excitation block enhanced the model's sensitivity to subtle facial manipulations without significantly increasing parameter complexity. The E-SAFE model has been shown to be superior in detecting subtle manipulations in deepfake images while maintaining parameter efficiency, thus potentially becoming a reliable solution for multimedia forensics.
Digital Transformation of Work Program Management in Higher Education Using a PDCA-Integrated Prototyping Approach Brilian Purnama Putra; Sugeng Purwantoro Edy Suranta G.S.; Istianah Muslim; Nina Fadilah Najwa
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3120

Abstract

Politeknik Caltex Riau is a higher education institution that manages various work programs to achieve its institutional vision and mission. The work program management process currently in place relies on Google Sheets as the primary platform, which has led to several issues, including inconsistencies in data entry, difficulties in delegating tasks, and a lack of harmonization among structural units. To address these challenges, a web-based Work Program Management Information System was developed using the Laravel framework and a prototyping development approach. The system is designed to support the complete Plan–Do–Check–Act (PDCA) cycle and to facilitate the digital planning, coordination, implementation, and evaluation of work programs. Based on the User Acceptance Test (UAT) results, the system achieved a 100% acceptance rate. Meanwhile, the Usability Testing produced average scores of 88% for usefulness, 85% for ease of use, 82.2% for ease of learning, and 84.4% for satisfaction. These results indicate that the developed system provides significant benefits to the institution by enhancing the efficiency and quality of work program management.
Optimized Skill Mastery Prediction for Adaptive Test Decision-Making Linda Perdana Wanti; Rujianto Eko Saputro; Fandy Setyo Hutomo; Muhammad Nur Faiz
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3171

Abstract

Adaptive testing systems require accurate and timely estimation of students' skill mastery levels to support decision-making in question selection and assessment flow determination. However, uncertainty in students' knowledge levels and variations in learning behavior often limit the performance of conventional predictive models. This study proposes an optimized predictive modeling framework for skill mastery to enhance decision-making quality in adaptive testing. The proposed framework integrates a machine learning-based skill mastery prediction model with an optimization mechanism to improve model accuracy and stability, while accommodating the sequential nature and uncertainty of student responses. Learning interaction data is used to dynamically model the development of skill mastery levels, which are then utilized as decision-support input in the adaptive testing system. The proposed predictive skill mastery model shows strong and consistent performance with an AUC value of 0.822, Average Precision of 0.868, accuracy of 0.757, and a precision balance of 0.834, recall of 0.788, and F1-score of 0.810, supported by well-calibrated probabilities and the ability to respond adaptively to student learning dynamics, making it suitable for use to support decision-making in adaptive test decision-making systems. The results of this study confirm the potential of integrating predictive analytics and optimization techniques in developing intelligent adaptive assessment systems.
Evaluating Cross-Language Structural Generalization of the Unified Abstract Syntax Tree Mardi Utomo; Ema Utami; Kusrini; Arief Setyanto
Journal of Innovation Information Technology and Application (JINITA) Vol 8 No 1 (2026): JINITA, June 2026
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v8i1.3191

Abstract

Cross-language code analysis requires syntax-aware representations that can reduce language-specific syntactic variation while preserving comparable program structure. Although unified Abstract Syntax Tree (AST) representations have been proposed, empirical evidence on their representation-level structural behavior across datasets and programming languages remains limited. This paper evaluates the structural generalization of a Unified AST representation as a schema-level abstraction, not as a parser-free or full semantic equivalence mechanism. The Unified AST schema is constructed from the CodeXGLUE code-to-text dataset covering Python, PHP, Ruby, Java, JavaScript, and Go. Its generalization is then examined on function-level aligned benchmarks from CodeXGLUE code-to-code translation (Java-C#) and multilingual HumanEval (Java, JavaScript, Go, Python, C++, and Rust). Tree Edit Distance (TED) similarity is used as the primary structural metric, while cosine similarity, BLEU, compression ratio, and identifier precision-recall are treated as auxiliary indicators of lexical similarity, reconstruction fidelity, compactness, and identifier retention. The results show an average TED similarity of 0.77 on CodeXGLUE code-to-code translation and 0.60 on HumanEval. These findings indicate that the Unified AST can preserve cross-language structural patterns under aligned benchmark assumptions, although it does not prove behavioral equivalence and remains dependent on language-specific parsing during AST extraction

Page 18 of 20 | Total Record : 191

 16   17   18   19   20 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol 8 No 1 (2026): JINITA, June 2026 Vol 7 No 2 (2025): JINITA, December 2025 Vol 7 No 1 (2025): JINITA, June 2025 Vol 6 No 2 (2024): JINITA, December 2024 Vol 6 No 1 (2024): JINITA, June 2024 Vol 5 No 2 (2023): JINITA, December 2023 Vol 5 No 1 (2023): JINITA, June 2023 Vol 4 No 2 (2022): JINITA, December 2022 Vol 4 No 1 (2022): JINITA, June 2022 Vol 4, No 1 (2022): JINITA, June 2022 Vol 3 No 1 (2021): JINITA, June 2021 Vol 3 (2021): JINITA, December 2021 Vol 2 No 2 (2020): JINITA, December 2020 Vol 2 No 1 (2020): JINITA, June 2020 Vol 1 No 1 (2019): JINITA, December 2019 More Issue