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
Kota yogyakarta,
Daerah istimewa yogyakarta
INDONESIA
International Journal of Advances in Intelligent Informatics
Published by Universitas Ahmad Dahlan
ISSN : 24426571     EISSN : 25483161     DOI : 10.26555
Core Subject : Science,
Computer Science & IT
International journal of advances in intelligent informatics (IJAIN) e-ISSN: 2442-6571 is a peer reviewed open-access journal published three times a year in English-language, provides scientists and engineers throughout the world for the exchange and dissemination of theoretical and practice-oriented papers dealing with advances in intelligent informatics. All the papers are refereed by two international reviewers, accepted papers will be available on line (free access), and no publication fee for authors.
Arjuna Subject : -
Articles 16 Documents
 1   2 
Search results for , issue "Vol 12, No 1 (2026): February 2026" : 16 Documents clear
A cascaded classification approach using transfer learning and feature engineering for improved breast cancer classification Ferkous, Chokri; Fadel, Ouissal; Kefali, Abderrahmane; Merouani, Hayet-Farida
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.1670

Abstract

The primary objective of this study is to design a cascaded classification framework that integrates deep-learning representations with handcrafted and clinical features to enhance the reliability and accuracy of breast cancer detection in mammographic screening. A multi-source mammography dataset comprising four databases was used to ensure diversity and reduce bias. The proposed system operates in two stages. In the first stage, transfer learning models (VGG16, ResNet50, and EfficientNet_B0) were evaluated using ROC-AUC, PR-AUC, calibration curves, and bootstrap confidence intervals. EfficientNet_B0, which achieved the best balance between discrimination and calibration, was selected as the feature extractor. In the second stage, the malignancy probability was combined with Haralick texture features, patient age, and breast density, and classified using SVM, Random Forest, MLP, Decision Tree, and Logistic Regression. Model robustness was verified through multi-run experiments (five random seeds) and subgroup analyses by age and density. Among the CNN models, EfficientNet_B0 yielded the best performance (accuracy = 0.9438, ROC-AUC = 0.944, PR-AUC = 0.960). In the second stage, although Random Forest achieved the highest accuracy (0.9556 ± 0.002), SVM obtained the highest mean ROC-AUC (0.980 ± 0.001) with stable accuracy (0.9539 ± 0.001) and the most significant p-values, indicating superior robustness and generalization. The proposed cascaded framework effectively combines deep, handcrafted, and clinical features to improve mammogram classification performance. The SVM-based model demonstrates strong calibration, stability, and subgroup consistency, highlighting its potential for deployment in computer-aided mammography screening systems that assist radiologists in early breast cancer detection.
Augmented haar cascade classifier for real-time ball detection in humanoid robots under dynamic environments Setyawan, Gembong Edhi; Widasari, Edita Rosana; Prasetio, Barlian Henryranu; Umar, Yasa Palaguna; Adipratama, Ivan Rafli
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2146

Abstract

This study proposes an Augmented Haar Cascade Classifier (AHCC) to enhance real-time ball detection for humanoid robots operating in dynamic environments. The method integrates Convex Hull mapping, HSV-based segmentation, and Hough Circle validation to overcome challenges such as fluctuating illumination, complex backgrounds, and partial occlusions. Experiments were conducted entirely on a CPU-only Intel NUC platform running ROS without GPU acceleration, using a dataset containing variations in lighting, orientation, scale, and background clutter. Compared with baseline models (standard Haar Cascade Classifier (HCC) and YOLOv5) the proposed AHCC achieved 97% accuracy, 83% recall, 97% precision, and an 89% F1-score, while requiring only 0.00849 s per frame with 8.97% memory usage. Although YOLOv5 reached 99% accuracy, it demanded higher computational resources (0.0344 s per frame, 22.3% memory usage), limiting its practicality for embedded robotic systems. The AHCC therefore offers an optimal balance between detection reliability and computational efficiency, outperforming traditional HCC and providing a lightweight alternative to GPU-dependent detectors such as Tiny-YOLO and MobileNet-SSD.
TUD-BISINDO: A new dataset and its recognition system using YOLO Raihan, Muhammad; Lestari, Aulia Ayu Dyah; Aulia, Suci; Hariyani, Yuli Sun; Maharani, Devira Anggi
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2329

Abstract

This study addresses the urgent need for digital inclusivity by developing a high-precision, real-time recognition system for Bahasa Isyarat Indonesia (BISINDO). The main new idea in this study is the creation of the Telkom University Database (TUD)-BISINDO, which is a strong and varied collection of data designed to fix the problems of current sign language databases, like not having enough different environments and camera angles. The TUD-BISINDO was created using 1,040 original images and added 780 more images to fix problems like differences in lighting, angles, and hand features that were often found in earlier datasets. The YOLOv8l model, improved with the AdamW optimizer and a flexible learning rate, performed exceptionally well with a mAP50 of 99.30% mAP50-95 of 85.40%, 99.80% precision, and 99.70% recall. These results demonstrate that the model significantly outperforms the previous YOLOv5 baseline across all primary metrics. The model has outstanding precision in recognizing real-time finger movements. However, complicated gestures, including the G and Z letters, require additional improvement. This research enhances sign language recognition technology, encouraging inclusion and improving accessibility for real-time communication. Future studies should focus on diversifying the dataset and maximizing performance in challenging conditions.
Comparative analysis of xception and svm for brain tumor classification on mri images Huda, Nurul; Ku-Mahamud, Ku Ruhana
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2016

Abstract

Brain tumor classification from magnetic resonance imaging (MRI) plays a critical role in supporting radiologists during diagnosis and treatment planning. However, many existing automated approaches employ limited preprocessing, single-stage transfer learning, or evaluation on a single dataset, which restricts robustness and clinical applicability. This study proposes an enhanced transfer-learning framework based on the Xception architecture for multiclass brain tumor classification and compares its performance with baseline models under identical experimental conditions. The framework integrates a comprehensive preprocessing pipeline consisting of normalization, adaptive noise filtering, contrast enhancement, and targeted data augmentation, together with a structured two-phase fine-tuning strategy. A total of 6,537 MRI images were used, employing five-fold cross-validation, independent testing, and validation on an additional benchmark dataset. The proposed model achieved a mean cross-validation accuracy of 0.8994 ± 0.089 and 99.06% accuracy, precision, and recall on the independent test set, demonstrating strong stability and generalization ability. Evaluation on the CE-MRI Figshare dataset further confirmed robustness, yielding 98.45% accuracy, 98% precision, and 98% recall. In contrast, when re-evaluated in the same experimental setting, baseline models performed considerably worse: the SVM classifier achieved 21.41% accuracy, and ResNet50 reached 75.27%, both substantially lower than Xception. Although higher accuracies for these models have been reported in prior studies under different conditions, the present findings highlight their limited generalization under unified evaluation. Overall, the proposed Xception-based framework provides a reliable and generalizable solution for automated brain tumor classification, with strong potential to support clinical workflows such as triage prioritization and second-opinion assistance.
Fine-tuned hyperparameter optimization for phishing website detection: insights into efficiency and performance Wahyudi, Rizki; Barkah, Azhari Shouni; Selamat, Siti Rahayu; Subarkah, Pungkas
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.1920

Abstract

The escalation of digital threats has made phishing-site identification a critical aspect of online protection. This study investigates how systematic hyperparameter adjustment through grid search influences both predictive precision and computational efficiency in phishing detection. Nine supervised classifiers from different algorithmic families were analyzed: tree-based models (DT, RF, GB, XGBoost), margin and distance-based learners (SVM, k-NN), probabilistic and neural approaches (NB, MLP), and a linear baseline using logistic regression (LR). Although machine learning (ML) approaches have demonstrated strong predictive capability, their reliability largely depends on precise parameter calibration. Through systematic exploration of parameter combinations, the grid-search approach identifies optimal settings for each model. Using the Kaggle phishing-URL dataset, tuned models achieved noticeable accuracy gains. DT, RF, and k-NN reached 99.1% accuracy with training times of 0.10 s, 1.55 s, and 0.01 s, respectively. MLP yielded 99.0% accuracy but required 2758 s, while SVM and LR achieved 97.8% and 92.9%. NB did the worst (62.7%). The results indicate that careful hyperparameter optimization enhances predictive ability, whereas model complexity heavily impacts runtime. This study’s novelty lies in a balanced assessment of accuracy and efficiency trade-offs, offering guidelines for selecting computationally efficient algorithms in practical phishing-detection systems.
Synergistic preprocessing approaches for improved time series analysis Pranolo, Andri; Pujiyanta, Ardi; Supriyanto, Supriyanto
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2321

Abstract

This paper systematically evaluates the performance of an LSTM baseline model, along with four smoothing augmentation methods (Kalman, Laplace, Moving Average, Savitzky-Golay), under different normalization strategies (Min-Max and Z-Score) for multivariate time-series forecasting. Experiments were conducted on six publicly available datasets (electricity consumption, energy consumption, sensor data, household energy, Indian electricity, and Brazilian temperature), and model performance was comprehensively compared using three metrics: MAPE, RMSE, and R². Results indicate that Laplace smoothing achieved the best performance across five datasets, effectively reducing errors while maintaining high fit quality, demonstrating its advantage in handling highly volatile and noisy time-series data. However, in some instances, Laplace smoothing, along with MA and SG methods, may produce an “over-smoothing” effect, causing forecasts to lose sensitivity to spike fluctuations. The choice of normalization strategy is equally critical: Min-Max is more suitable for data with stable distributions, while Z-Score demonstrates greater advantages for data with large numerical ranges and significant volatility. Notably, in temperature datasets with small sample sizes and high volatility, complex smoothing methods actually degraded performance, making the baseline LSTM + Z-Score the optimal choice. However, the LSTM-Laplace model with Min-Max normalization achieves the best performance among the models. Overall, the study concludes that improving prediction performance relies not only on model architecture but also on optimizing data scale, distribution characteristics, and preprocessing strategies.
Single-input and multi-input local binary pattern classification Manga, Abdul Rachman; Handayani, Anik Nur; Herwanto, Heru Wahyu; Asmara, Rosa Andrie; Raja, Roesman Ridwan
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2183

Abstract

Identification and classification of species are crucial for maintaining genetic diversity and supporting sustainable agricultural practices. The Toraja Buffalo, a unique type of buffalo in Indonesia, holds high cultural and economic value. Accurate classification of this species is essential to preserving genetic resources and improving breeding programs. Previous studies using single classification methods have shown limitations in complex cases such as the Toraja Buffalo, which has numerous physiological characteristics such as body size, head, horns, tail, and eyes. The purpose of this study is to evaluate and compare the performance of single-classification and multi-category methods for identifying Toraja Buffalo. Several algorithms, including K-Nearest Neighbors (K-NN), Random Forest, Support Vector Machine (SVM), and Naive Bayes, were tested using Local Binary Pattern (LBP) for feature extraction. Decision Tree and others were observed, showing 85.83% accuracy in single-input, while multi-input accuracy reached 92.08%. The multi-input approach consistently improved performance across all algorithms. Multi-input classifiers significantly outperformed single-feature methods, with Random Forest being the most efficient algorithm. Future research could incorporate additional variables such as skin color or genetic profiles to further enhance accuracy.
Collaborative filtering-based group recommender system using sparse autoencoder Bahar, Musthafa Zaki; Baizal, Zinke Abdurahman
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.1702

Abstract

The development of technology makes the distribution of information easier and faster, but leads to information overload. A recommender system is one tool to overcome information overload, while the collaborative filtering (CF) paradigm is a widely used approach in recommender systems. The recommender system generally focuses on individual recommendations, but in real conditions, recommendations for a group are often needed, for example, when we want to listen to music with friends, or we plan a vacation with family. Many prior studies have used the CF paradigm with matrix factorization to build group recommender systems. Matrix factorization has been shown to alleviate the sparsity problem; however, it does not fully resolve it. Therefore, we propose an approach that uses a sparse autoencoder to address this sparsity issue. We chose the sparse autoencoder because it can effectively capture latent patterns in sparse data by learning a compressed representation while retaining important features crucial for accurate recommendations. We built a group recommender system with three different group sizes and aggregation approaches. For evaluation, we use the root-mean-square error (RMSE) and the mean absolute error (MAE). Test results indicate that the sparse autoencoder outperforms matrix factorization in terms of RMSE and MAE. This study improves group recommender systems by addressing data sparsity using a sparse autoencoder. The proposed approach enhances recommendation accuracy compared to traditional matrix factorization methods.
Underwater image enhancement with fuzzy histogram equalization and adaptive color correction Suharyanto, Suharyanto; Andono, Pulung Nurtantio; Fanani, Ahmad Zainul; Pujiono, Pujiono
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2174

Abstract

Marine exploration continues to increase as new technologies, such as computer vision implemented in underwater vehicles and robots, develop. Identifying underwater objects is challenging due to environmental conditions, including poor lighting and color absorption in the viewed image. Underwater image enhancement has been widely applied to overcome these obstacles. Therefore, this study presents a new workflow for improving the quality of underwater images. A combination of the fuzzy histogram equalization (FHE) and adaptive color correction (ACC) methods is used to increase contrast and restore absorbed colors. This study proposes combining FHE and ACC to improve underwater image quality, using the FHE method with the FHEACC method. The results of the UIQM and ENTROPY metrics obtained the highest values, while UCIQE ranked third. This shows that the image quality improved using the FHEACC combination method is objectively better than that achieved with the HE, AHE, CLAHE, FHE, IBLA, RCP, and UDCP methods, especially in maintaining color balance. This research can introduce a new workflow to improve the quality of underwater images by combining Fuzzy Histogram Equalization and Adaptive Color Correction methods, thereby supporting the optimization of underwater image identification systems in wild environments using computer vision technology.
A comprehensive comparative analysis of chicken meat classification techniques through machine learning models Anraeni, Siska; Lahuddin, Harlinda; Ramdaniah, Ramdaniah; Melani, Erika Riski; Amalia, Andi Cici; Amaliah, Tazkirah
International Journal of Advances in Intelligent Informatics Vol 12, No 1 (2026): February 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i1.2014

Abstract

This study develops a digital image processing technique to distinguish between fresh and rotten chicken. Chicken freshness has a significant impact on public health and industry sustainability. This study uses a multi-stage approach including data acquisition, preprocessing, feature extraction, and classification. A total of 1,000 chicken images were obtained, consisting of 800 images for training and 200 images for testing, with a proportion of 80:20. Feature extraction was performed using a combination of the HSI (Hue, Saturation, Intensity) color model to capture the color characteristics of chicken and the Local Binary Pattern (LBP) to extract texture information. Classification was performed using the K-Nearest Neighbor (KNN) algorithm with various K values and distance metrics. The experimental results show that the combination of color and texture features provides higher accuracy than using either feature alone. The best model using HSI and LBP feature extraction with K = 1 and K = 3 in the Euclidean distance metric achieved the highest accuracy of 95.4%. With a promising level of accuracy, this method can be applied in automated inspections in the poultry supply chain, improving food safety and helping consumers make better purchasing decisions. However, the main challenge in this study is the variation in lighting during image capture, which causes the fresh and rotten chicken feature values to overlap, thus hindering perfect classification.

Page 1 of 2 | Total Record : 16

 1   2 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue Vol 12, No 1 (2026): February 2026 Vol 11, No 4 (2025): November 2025 Vol 11, No 3 (2025): August 2025 Vol 11, No 2 (2025): May 2025 Vol 11, No 1 (2025): February 2025 Vol 10, No 3 (2024): August 2024 Vol 10, No 2 (2024): May 2024 Vol 10, No 1 (2024): February 2024 Vol 9, No 3 (2023): November 2023 Vol 9, No 2 (2023): July 2023 Vol 9, No 1 (2023): March 2023 Vol 8, No 3 (2022): November 2022 Vol 8, No 2 (2022): July 2022 Vol 8, No 1 (2022): March 2022 Vol 7, No 3 (2021): November 2021 (Issue in Progress) Vol 7, No 3 (2021): November 2021 Vol 7, No 2 (2021): July 2021 (Issue in Progess) Vol 7, No 2 (2021): July 2021 Vol 7, No 1 (2021): March 2021 Vol 6, No 3 (2020): November 2020 Vol 6, No 2 (2020): July 2020 Vol 6, No 1 (2020): March 2020 Vol 5, No 3 (2019): November 2019 Vol 5, No 2 (2019): July 2019 Vol 5, No 1 (2019): March 2019 Vol 4, No 3 (2018): November 2018 Vol 4, No 2 (2018): July 2018 Vol 4, No 1 (2018): March 2018 Vol 3, No 3 (2017): November 2017 Vol 3, No 2 (2017): July 2017 Vol 3, No 1 (2017): March 2017 Vol 2, No 3 (2016): November 2016 Vol 2, No 2 (2016): July 2016 Vol 2, No 1 (2016): March 2016 Vol 1, No 3 (2015): November 2015 Vol 1, No 2 (2015): July 2015 Vol 1, No 1 (2015): March 2015 More Issue