cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
,
INDONESIA
JOURNAL OF APPLIED INFORMATICS AND COMPUTING
Published by Politeknik Negeri Batam
ISSN : -     EISSN : 25486861     DOI : 10.3087
Core Subject : Science,
Computer Science & IT
Journal of Applied Informatics and Computing (JAIC) Volume 2, Nomor 1, Juli 2018. Berisi tulisan yang diangkat dari hasil penelitian di bidang Teknologi Informatika dan Komputer Terapan dengan e-ISSN: 2548-9828. Terdapat 3 artikel yang telah ditelaah secara substansial oleh tim editorial dan reviewer.
Arjuna Subject : -
Articles 1,006 Documents
 84   85   86   87   88 
Comparative Analysis of Convolutional Neural Network Architectures in Pneumonia Detection Indah Putianik; Wise Herowati
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12738

Abstract

The lungs are one of the important organs for humans. One of the lung diseases that can potentially cause death is pneumonia. This study compares the performance of the Convolutional Neural Network (CNN), ResNet-50, and MobileNetV2 models in detecting pneumonia using chest X-ray images. The dataset used was obtained from Kaggle with a total of 2,000 chest X-ray images, which were divided into 70% training data, 15% validation data, and 15% test data. All images were resized to 256×256 pixels. The experiment was carried out 10 times with 20 training epochs and a batch size of 32. The results showed that the CNN model obtained an average accuracy of 95.76%, precision of 95.39%, recall of 96.20%, and F1-score of 95.77%. The ResNet-50 model produced an average accuracy of 96.99%, precision of 99.79%, recall of 94.20%, and F1-score of 96.90%. Meanwhile, MobileNetV2 achieved an average accuracy of 97.33%, precision of 97.23%, recall of 98.13%, and F1-score of 97.66%. ResNet-50 was better at minimizing False Positives, while MobileNetV2 was better at minimizing False Negatives. Overall, all three models performed well in detecting pneumonia, although there were variations in the confusion matrix results across experiments.
VGG16 Transfer Learning for Bone Fracture Classification Using X-Ray Images Megan Febriana Putri Johana; Christy Atika Sari
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12739

Abstract

Bone fracture is one of the most common injury conditions and requires a fast and accurate diagnosis process to assist optimal medical treatment. Examination using X-Ray images is the main method in identifying bone fractures, but the process of interpreting radiographic images has challenges, especially in Multi-class classification with similar fracture characteristics. This study aims to implement a transfer learning method based on the VGG16 architecture for Multi-class classification of bone fractures using X-Ray images. The dataset used consists of 11 classes, namely Avulsion Fracture, Comminuted Fracture, Fracture Dislocation, Greenstick Fracture, Hairline Fracture, Impacted Fracture, Longitudinal Fracture, Oblique Fracture, Pathological Fracture, Spiral Fracture, and Normal. The preprocessing stage includes resizing the image to 256 × 256 pixels, RGB conversion, VGG16 preprocessing, and data augmentation to increase the variety of the dataset. The model was built using pretrained VGG16 as a feature extractor with the addition of GlobalAveragePooling2D, Dense layer, BatchNormalization, and Dropout and fine-tuning was performed on several final layers. The evaluation results showed that the model obtained an accuracy of 98.40%, a macro precision of 97.95%, and a macro recall of 97.95%. In addition, most classes obtained accuracy values close to 100%. The results showed that the application of VGG16-based transfer learning was able to provide excellent classification performance on X-Ray images of bone fractures and was effective in improving the model's generalization ability in multi-class classification of medical images.
Comprehensive Diabetes Risk Prediction Using BRFSS Data: Performance, Explainability, Fairness, and Calibration Virzan Pasa Nugraha; Agung Febrian
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12740

Abstract

This study aims to develop and evaluate machine learning models for diabetes risk prediction using a comprehensive approach that considers performance, interpretability, fairness, and calibration aspects. The research employs several classification algorithms, including Logistic Regression, Random Forest, XGBoost, and CatBoost, using the BRFSS dataset. The models are evaluated using multiple metrics, including Accuracy, Balanced Accuracy, Precision, Recall, F1-Score, ROC-AUC, Precision-Recall AUC (PR-AUC), Matthews Correlation Coefficient (MCC), and Brier Score. Explainability analysis is conducted using SHAP to understand feature contributions, while fairness and calibration analyses are performed to assess model reliability and bias across demographic groups. The results show that CatBoost achieves the best overall performance, with the highest ROC-AUC and Recall, as well as the lowest Brier Score, indicating better predictive capability and calibration. Explainability analysis reveals that GenHlth, BMI, and Age are the most influential features, while fairness analysis indicates potential disparities across certain age groups. Furthermore, ablation and misclassification analyses highlight key features and areas for model improvement. Overall, this study demonstrates that integrating performance evaluation with explainability and fairness analysis can produce more reliable and interpretable predictive models for healthcare applications.
Knowledge Discovery of AI Usage Dependency Patterns in Learning Activities Using Random Forest, XGBoost, Logistic Regression with SHAP-Based Interpretation Fidela Tertia Alfino; Puti Chalisa Wardhana; A. Salwa Aurelya Putri; Athiyyah Nuha Rotifa; Ken Ditha Tania; Ahmad Rifai; Dedy Kurniawan
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12745

Abstract

The increasing use of Artificial Intelligence (AI) in education has influenced various learning activities. However, excessive AI usage has the potential to create dependency patterns that may affect students’ learning independence and critical thinking abilities. This study aims to analyze patterns of AI usage dependency in learning activities using a machine learning approach and to interpret the factors influencing such dependency. The analysis was conducted using a publicly available dataset representing usage intensity, session duration, AI assistance level, repeated usage behavior, and students’ academic characteristics. The research stages consisted of data preprocessing, categorical variable encoding, feature engineering, the construction of the Knowledge Dependency Level variable, class imbalance handling using SMOTE, and model evaluation using Stratified 5-Fold Cross Validation. The dataset was divided into 80% training data and 20% testing data, then modeled using Logistic Regression, Random Forest, and XGBoost. The results showed that XGBoost achieved the best performance with an accuracy of 0.6845, precision of 0.7288, recall of 0.6845, F1-score of 0.7028, and an AUC value of 0.860, indicating better discrimination capability compared to Random Forest and Logistic Regression. To support the knowledge discovery process, an interpretative analysis using SHAP was conducted to identify the contribution of each feature to the classification results. The interpretation revealed that SatisfactionRating was the most dominant feature influencing the prediction of AI usage dependency levels, followed by FinalOutcome, while academic factors such as StudentLevel and Discipline contributed relatively less. These findings transform previously implicit AI usage behavior patterns into explicit knowledge.
Classification Model Evaluation of Public Perception Regarding Nusantara Capital Relocation Galih Sabilah Rosat; Nova Rijati
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12747

Abstract

The relocation of the Capital City of Nusantara (IKN) has triggered extensive public opinion dynamics on the X (Twitter) social media platform. This study aims to empirically map public sentiment and evaluate the comparative performance of classification machine learning algorithms: Support Vector Machine (SVM), Logistic Regression (LR), and Multinomial Naïve Bayes (MNB). The dataset comprises 2,257 unique texts spanning the 2024-2025 period, automatically categorized using a lexicon-based approach (InSet dictionary). The technical novelty of this research lies in the extraction of TF-IDF N-gram (1, 2) features, the mitigation of extreme class imbalance via the Synthetic Minority Over-sampling Technique (SMOTE) encapsulated within a cross-validation pipeline, and systematic hyperparameter optimization using GridSearchCV. The labeling results revealed a predominance of negative sentiment at 77.3% (1,745 data) compared to positive sentiment at 22,7% (512 data), reflecting high public scrutiny and skepticism. Based on the model evaluation, MNB with parameter optimization (α=0.1) emerged as the most robust model, achieving an accuracy of 80.5% and a Macro Precision of 72.3%. The performance of MNB proved capable of outperforming the architectures of LR and SVM, which struggled with multicollinearity and underfitting anomalies in high-dimensional spaces. These findings emphasize that the integration of SMOTE and the MNB algorithm is crucial in overcoming majority class bias and calibrating the limitations of static lexicons to precisely map the polarity of public opinion.
A Robust Voting Ensemble Framework for Predicting Thermal Stability in Zn-Based Metal–Organic Frameworks Taufiqul Umam; Harun Al Azies; Muhamad Akrom; Ananta Surya Pratama; Muhammad Diva Irnanda
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12748

Abstract

The prediction of thermal stability (TS) in zinc-based metal–organic frameworks (Zn-MOFs) is often challenged by experimental cost and distributional heterogeneity in materials datasets. This study proposes a median-based robust voting ensemble to model the TS of 151 Zn-MOF samples using four structural descriptors, nN, nZn, Het, and Lig. The framework integrates five robust linear estimators and is benchmarked against a linear kernel Support Vector Regression (SVR) model to evaluate predictive stability and generalization performance. The proposed ensemble demonstrates superior test performance (R² = 0.9986; RMSE = 0.0023) compared to SVR (R² = 0.9492; RMSE = 0.0213), indicating enhanced robustness under heterogeneous data conditions. Feature importance analysis identifies nitrogen coordination density and heteroatomic environment as the dominant contributors to TS prediction, while zinc center quantity and ligand topology exhibit comparatively minor influence. These findings confirm that median-based robust aggregation improves predictive reliability and provides chemically interpretable insight, offering a data-driven approach for the rational design and screening of thermally stable Zn-MOF materials.
Integrated Emergency Communication System for Disaster Areas Using Long Range Mahadika Luqman; Muhammad Fikry; Yesy Afrillia
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12753

Abstract

Natural disasters disrupt communication infrastructure, hindering emergency response coordination. This study designs and evaluates an integrated emergency communication system combining LoRa for transmission, GPS for geolocation, and BLE for alternative interface. The system comprises a Field Device with a 9-state finite-state machine, a Beacon Network forming a linear multi-hop relay chain with heartbeat-based node failure detection, and a Headquarter Device connected to the Blynk platform for monitoring and notifications. A custom binary protocol with 8 message types uses packed structures. All performance was evaluated in urban area, except maximum communication direct range in urban area and rural area. PDR achieves 100% up to 1,000 m Line-of-Sight with an average end-to-end latency of 1.02 s. A single beacon relay extends communication range to 2000 m compared to maximum direct communication range, 1288 m in rural area and 1044 m in urban campus area. Outdoor GPS accuracy measures 0.945 m, while indoor accuracy 28.68 m due to building attenuation. The system successfully detected motion >5 m with 100% sensitivity within 5 s. Usability testing average completion times of 14.84 s via physical interface and 23.85 s via mobile application. BLE range reaches 16 m outdoors and 11 m indoors. Operational durations were 4.32 h for the Field Device, 9.18 h for the Beacon Network, and 8.03 h for the Headquarter Device, falling short of the 12-hour target, necessitating aggressive GPS duty cycling and Wi Fi sleep modes. This study evaluates three critical factors for emergency response: network determinism, payload efficiency, and power autonomy.
Comparative Analysis of Machine Learning and IndoBERT Models for Sentiment Analysis of YouTube Comments on the Free Nutritious Meals Program Irvan Theo Shandy; Ucta Pradema Sanjaya
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12754

Abstract

The Free Nutritious Meals Program has become one of the most widely discussed public policies in Indonesia and has generated various public responses on social media, particularly YouTube. Public comments on YouTube can be utilized as a valuable data source to understand public sentiment toward the implementation of the program. Therefore, this study aims to analyze and compare the performance of several classification algorithms in sentiment analysis of YouTube comments related to the Free Nutritious Meals Program. The dataset used in this study was obtained through a crawling process on one of Raymond Chin’s YouTube videos discussing the MBG program. A total of 903 comments were collected, and after the preprocessing stage, 401 comments were selected for further analysis. The preprocessing steps included cleaning, normalization, tokenization, stopword removal, and stemming. Furthermore, the text data were transformed using the TF-IDF weighting method. This study compared several classification algorithms, namely Random Forest, Gradient Boosting, Support Vector Machine (SVM), XGBoost, Multinomial Naïve Bayes, IndoBERT, and LightGBM. Model evaluation was conducted using confusion matrix analysis and performance metrics consisting of accuracy, precision, recall, and F1-score. The experimental results show that the Random Forest algorithm achieved the best performance with an accuracy of 0.9672, precision of 0.9683, recall of 0.9672, and F1-score of 0.9620. However, the confusion matrix analysis indicates that the model tends to be biased toward the positive sentiment class due to the imbalance in sentiment distribution within the dataset. In addition, the relatively small dataset and the use of comments from a single YouTube source may affect the generalization of the model results. Based on these findings, Random Forest can be considered the most effective algorithm for sentiment classification in this study. The results of this research are expected to provide insights into public perceptions regarding the MBG program and serve as evaluation material for policymakers in improving the implementation of public nutrition programs in Indonesia.
Enhancing Inflation Forecasting in Indonesia Using N-BEATSx with Exogenous Factors Talitha Adilla Fujisai Panglima Putri; Mohammad Idhom; Muhammad Nasrudin
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12759

Abstract

Accurate inflation forecasting is crucial for economic stability and effective policymaking, particularly in emerging economies such as Indonesia, where monetary policy, global commodity markets, exchange rate fluctuations, and recurring religious seasonal events simultaneously influence price dynamics. This study proposes an inflation forecasting framework using the N-BEATSx (Neural Basis Expansion Analysis for Time Series with Exogenous Variables) deep learning model, incorporating macroeconomic variables, global oil prices, BI Rate, and the USD/IDR exchange rate, alongside Ramadan and Eid al-Fitr calendar dummy variables as exogenous inputs. The dataset comprises 153 monthly observations spanning January 2013 to September 2025, split into training, validation, and test sets, with a forecasting horizon of six months. The N-BEATSx model is benchmarked against SARIMAX, LSTM, and Prophet. Results on the test set show that N-BEATSx achieves competitive performance (RMSE 0.0067, MAE 0.0058, SMAPE 51.77%) outperforming SARIMAX (RMSE 0.0297, MAE 0.0266) and LSTM (RMSE 0.0098, MAE 0.0084). Although Prophet yields marginally lower absolute errors, the MAE gap is minimal (0.0002), while N-BEATSx offers superior interpretability through an explicit decomposition of forecasts into trend, seasonality, and exogenous components. Component decomposition analysis reveals that macroeconomic exogenous variables dominate the forecast output, confirming their theoretical relevance as inflation drivers. Six-month-ahead forecasts project inflation in the range of 2.47% - 3.54% for October 2025 to March 2026, approaching Bank Indonesia’s upper target corridor, suggesting the need for preemptive monetary policy measures.
Implementation and Performance Analysis of Mel Frequency Cepstral Coefficient Features in Dangdut Music Sub-Genre Classification I Nyoman Surya Jaya; Tria Hikmah Fratiwi; I Gede Harsemadi; I Dewa Made Dharma Putra Santika; Ni Putu Nanda Maharani; Putu Andiny Julia Putri Rapayana
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12760

Abstract

This study investigates the classification of dangdut music sub-genres using Mel-Frequency Cepstral Coefficients (MFCC) and machine learning approaches. The objective is to evaluate the effectiveness of MFCC in representing audio characteristics and to compare the performance of several classification algorithms, including K-Nearest Neighbor (K-NN), hybrid K-NN optimized with Genetic Algorithm (GA), Support Vector Machine (SVM), and Decision Tree. The dataset consists of 730 audio samples with a duration of 30 seconds each, categorized into three sub-genres: classic dangdut, rock dangdut, and koplo dangdut. The research process includes audio segmentation, extraction of 13 MFCC coefficients, data normalization, train-test splitting (70:30), and performance evaluation using accuracy, precision, recall, F1-score, confusion matrix, and cross-validation. The results indicate that MFCC provides discriminative feature representations, as demonstrated by improved cluster separation in PCA after normalization. Among the evaluated models, hybrid K-NN with GA achieved the highest accuracy of 98.90%, outperforming conventional K-NN, SVM, and Decision Tree. Confusion matrix analysis showed that most samples were correctly classified, with only minor misclassifications between sub-genres sharing similar audio characteristics. Furthermore, consistently high cross-validation accuracies and low standard deviation values confirmed good generalization capability and suggested the absence of significant overfitting. Overall, the findings demonstrate that MFCC is an effective feature for dangdut music sub-genre classification, while normalization and GA-based optimization significantly improve classification performance and model robustness.

Page 86 of 101 | Total Record : 1006

 84   85   86   87   88 

Filter by Year

2017 2026


Reset
Filter By Issues
All Issue Vol. 10 No. 4 (2026): August 2026 Vol. 10 No. 3 (2026): June 2026 Vol. 10 No. 2 (2026): April 2026 Vol. 10 No. 1 (2026): February 2026 Vol. 9 No. 6 (2025): December 2025 Vol. 9 No. 5 (2025): October 2025 Vol. 9 No. 4 (2025): August 2025 Vol. 9 No. 3 (2025): June 2025 Vol. 9 No. 2 (2025): April 2025 Vol. 9 No. 1 (2025): February 2025 Vol. 8 No. 2 (2024): December 2024 Vol. 8 No. 1 (2024): July 2024 Vol. 7 No. 2 (2023): December 2023 Vol. 7 No. 1 (2023): July 2023 Vol. 6 No. 2 (2022): December 2022 Vol 6 No 1 (2022): July 2022 Vol. 5 No. 2 (2021): December 2021 Vol 5 No 2 (2021): December 2021 Vol 5 No 1 (2021): July 2021 Vol. 5 No. 1 (2021): July 2021 Vol. 4 No. 2 (2020): Desember 2020 Vol 4 No 2 (2020): Desember 2020 Vol 4 No 1 (2020): Juli 2020 Vol 3 No 2 (2019): Desember 2019 Vol 3 No 1 (2019): Juli 2019 Vol 2 No 2 (2018): Desember 2018 Vol. 2 No. 1 (2018): Juli 2018 Vol 2 No 1 (2018): Juli 2018 Vol 1 No 2 (2017): Desember 2017 Vol 1 No 1 (2017): Juli 2017 More Issue