cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Imam Much Ibnu Subroto
Contact Email
imam@unissula.ac.id
Phone
-
Journal Mail Official
ijai@iaesjournal.com
Editorial Address
-
Location
Kota yogyakarta,
Daerah istimewa yogyakarta
INDONESIA
IAES International Journal of Artificial Intelligence (IJ-AI)
Published by Institute of Advanced Engineering and Science
ISSN : 20894872     EISSN : 22528938     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Engineering
IAES International Journal of Artificial Intelligence (IJ-AI) publishes articles in the field of artificial intelligence (AI). The scope covers all artificial intelligence area and its application in the following topics: neural networks; fuzzy logic; simulated biological evolution algorithms (like genetic algorithm, ant colony optimization, etc); reasoning and evolution; intelligence applications; computer vision and speech understanding; multimedia and cognitive informatics, data mining and machine learning tools, heuristic and AI planning strategies and tools, computational theories of learning; technology and computing (like particle swarm optimization); intelligent system architectures; knowledge representation; bioinformatics; natural language processing; multiagent systems; etc.
Arjuna Subject : -
Articles 2,143 Documents
 205   206   207   208   209 
Fine-tuning OpenAI’s Whisper model for Kazakh speech recognition Bakdaulet Kynabay; Aimoldir Aldabergen; Shirali Kadyrov; Ardak Shalkarbay-uly
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4053-4060

Abstract

Speech recognition remains a challenging task, especially for low-resource languages such as Kazakh. This study explores the effectiveness of fine-tuning OpenAI’s Whisper model on Kazakh speech data for improving speech recognition accuracy. The Kazakh speech corpus (KSC) is used as a benchmark dataset and to conduct experiments to evaluate how the adapted Whisper framework measures up against current leading speech recognition models, including deep neural network (DNN)-hidden Markov model (HMM), end-to-end long short-term memory (E2E-LSTM), and end-to-end transformer (E2E-transformer). Research outcomes indicate that the Whisper model outperforms all baselines after fine-tuning, showing a result with a word error rate (WER) of 12.5%, a character error rate (CER) of 3.15%, a character-level F1-score (chrF) of 93.72, and a bilingual evaluation understudy (BLEU) score of 76.05. These findings suggest that the Whisper model that is fine-tuned on Kazakh audio data can be a promising approach for improving Kazakh speech recognition accuracy. This work advances the development of speech recognition for the Kazakh language by demonstrating the effectiveness of transfer learning in low-resource languages. Subsequent research must prioritize collecting larger and more diverse speech data for Kazakh, including noisy and accented speech, to further improve speech recognition accuracy.
Feature-engineered stacking and AutoGluon framework for deep learning-based autism detection Savita Rani; Chandra Shakher Tyagi; Devendra Kumar
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4260-4272

Abstract

Autism spectrum disorder (ASD) is a brain development disorder that distracts a person’s perceptions and social interactions with others. Current studies emphasize conventional machine learning and typical ensemble methods and little focus have been placed on merging automated machine learning (AutoML) systems with stacking ensemble structures. To bridge this gap, this study evaluates stacking ensembles with an artificial neural network (ANN)-based meta-classifier and the AutoGluon framework. For enhancing model stability and generalizability across various ASD datasets, systematic RandomizedSearchCV along with 10-fold stratified cross-validation is utilized. The stacking model outperformed the AutoGluon framework and all individual classifiers on dataset-I, achieving 99.50% accuracy, 99.53% precision, 99.53% recall, and a 99.97% receiver operating characteristic-area under the curve (ROC-AUC). With 97.28% accuracy, precision 96.33%, 95.38 F1-score, and a 99.76% ROC-AUC on dataset-II, the model demonstrated great generalization and continuously outperformed traditional ensemble and boosting techniques. Stacking model perform well on both datasets and maintain consistence, whereas AutoGluon give competitive performance on dataset-1 and it exhibited a notable decline on dataset-II. K-fold internal and external cross-validation were used to access model’s generalization and consistency. The outcomes of stacking ensemble model indicate that model offers higher accuracy, greater resilience, better consistency across difference datasets and offers a robust approach to early ASD prediction.
LLM-FS-Agent: a deliberative role-based large language model architecture for transparent feature selection Mohamed Bal-Ghaoui; Fayssal Sabri
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4026-4038

Abstract

High-dimensional data can reduce interpretability and increase computational cost in machine learning (ML) pipelines. Large language models (LLMs) offer a promising approach for dimensionality reduction through feature selection (FS), but existing LLM-based methods lack structured reasoning and do not provide clear justifications for their choices. This study propose LLM-FS-Agent, a multi-agent architecture where specialized LLM roles collaborate through a deliberative process to evaluate and rank features. Each selection decision is accompanied by an explicit rationale. We tested our method on the CIC-DIAD 2024 internet of things (IoT) intrusion detection dataset and compared it against LLM-select and traditional techniques like principal component analysis (PCA) across multiple subset sizes. LLM-FS-Agent achieves comparable or better accuracy while reducing classifier training time by 46% on average (statistically significant, 0.094 s, p = 0.028 for extreme gradient boosting (XGBoost)). These results show that multi-agent deliberation improves transparency and computational efficiency in FS.
Hybrid feature selection strategies for strong network intrusion detection Josephine Rathinam; Anandaraj Shanthi Pichandi
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4366-4380

Abstract

High dimensional network traffic datasets pose great challenges to network intrusion detection systems (NIDS) due to redundant and irrelevant features, which decrease detection accuracy and increase the computational cost. Traditional feature selection methods often cannot achieve a balance between classification performance and processing efficiency. This paper proposes hybrid feature selection for network intrusion detection systems (HybFS-NIDS), a hybrid feature selection framework using filter- and wrapper-based optimization methods to detect the most significant intrusion detection features to do this. The proposed model was evaluated on the CIC-IDS2017 dataset using performance metrics such as accuracy, precision, recall, F1-score, and Matthew’s correlation coefficient (MCC). The experimental results selected 18 optimal features from the initial 78 features of the dataset and achieved 90.68% accuracy, 91.74% precision, 90.64% recall, 91.19% F1-score, and an MCC of 0.9148. The proposed method also achieved a 41.5% reduction in training time compared with the full-feature baseline model while improving the overall classification performance. The results show that HybFS-NIDS can successfully improve the efficiency of intrusion detection and fit for real-time cybersecurity applications with a large amount of network traffic data.
A feature-enhanced deep learning model for multi-class paddy ailment prediction in smart farming Santhosh Byramangala Jayaramaiah; Balarengadurai Chinnaiah; Boddu Ravi Prasad
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4470-4481

Abstract

Paddy is an essential staple food crop eaten by more than 50% of the world population, particularly in Asia, where countries like China and India account for over 50% of both production and consumption. However, paddy cultivation is frequently threatened by numerous ailments that badly impact crop production and quality. Normal ailment identification approaches are often manual, time-consuming, and error-prone. To address these challenges, deep learning (DL)-based approaches have been widely adopted for automated paddy ailment detection. While these models have achieved commendable accuracy, many fail to focus on effective feature extraction, which is vital for accurate classify across diverse ailment categories. Hence, this study proposes an advanced convolutional neural network (CNN) based approach associated with convolutional block attention module (CBAM) for enhancing feature representation by focusing on both spatial and channel-wise importance. The approach was trained and executed on the widely used “paddy doctor” dataset, covering 2-class, 10-class, and 13-class classify scenarios. The proposed CNN-CBAM achieved outcomes with 99.75%, 99.93%, and 99.92% accuracy, respectively, outperforming existing models. These outcomes show the model’s effectiveness in capturing discriminative features and improving classify performance. This model contributes a foundation for future extensions in precision agriculture, including pest detection and real-time monitoring.
Deep reinforcement learning-based credit risk prediction model for multi-attribute data Al-Khowarizmi Al-Khowarizmi; Ferry Fachrizal; Romi Fadillah Rahmat; Arif Ridho Lubis; Hariyadi Hariyadi
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4624-4634

Abstract

This paper aims to develop a multi-attribute data-based credit risk prediction model using the deep reinforcement learning (DRL) approach. In the context of creditworthiness evaluation, a model is needed that is not only accurate but also adaptive to complex data dynamics and risk patterns. This study uses a dataset of 5,000 entries that have gone through a preprocessing process, with features such as credit score, income, age, loan amount, and employment history. Initial correlation analysis through heatmap visualization shows that credit score has a strong negative relationship with credit risk, while other features show weak to moderate correlations. The evaluation of model performance is shown through the area under the curve (AUC) value on the receiver operating characteristic (ROC) curve of 0.88, which indicates excellent discrimination ability. The final evaluation results show that the DRL model has an accuracy of 91%, precision of 89%, recall of 90%, and F1-score of 89.5%, indicating that the model not only has high accuracy but is also balanced in its sensitivity and specificity. Thus, the application of DRL has proven effective in predicting credit risk adaptively and accurately, and can be implemented as an intelligent solution in an automated credit decision-making system based on multi-attribute data.
Interpretable machine learning framework for congestive heart failure risk prediction Sakchai Tangprasert; Nalinpat Bhumpenpein; Siranee Nuchitprasitchai; Yuenyong Nilsiam
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4345-4355

Abstract

Congestive heart failure (CHF) is a leading cause of morbidity and mortality worldwide and is influenced by multiple clinical risk factors, including age, sex, blood pressure, cholesterol levels, and electrocardiographic characteristics. This study employs publicly available heart disease datasets compiled from populations in Cleveland, Hungary, Switzerland, and Long Beach repositories (2015–2020) and analyzes 11 predictor attributes and one binary outcome using the cross-industry standard process for data mining (CRISP-DM) framework. Four supervised machine learning (ML) models—decision tree, random forest, gradient boosted trees, and naïve Bayes—were developed and evaluated using confusion matrix–based performance metrics, with results showing that the decision tree model offers balanced predictive performance and clear interpretability, making it suitable for heart failure risk prediction. The findings demonstrate the potential of ML approaches to support clinical risk assessment and enhance decision-making in heart failure management.
Fine-tuning language models for social sentiment: insights from startup user feedback Li-Hua Li; Agus Cahyo Nugroho
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4432-4446

Abstract

Startup growth and sustainability are heavily impacted by user comments. This research has done a comparative study between large language models (LLMs), small language models (SLMs), and traditional machine learning (ML) methods in classifying user comments on startups. The performances of bidirectional encoder representations from transformers (BERT), Indonesian bidirectional encoder representations from transformers (IndoBERT), distilled bidirectional encoder representations from transformers (DistilBERT), and ML models; logistic regression, support vector machine (SVM), random forest, and extreme gradient boosting (XGBoost) were evaluated with a diverse dataset and beyond extensive preprocessing. Out of all the models this research trained, IndoBERT outperformed with an accuracy and recall 90.99% and F1-score of 89.94%, due to its intermediate pretraining for Indonesian language task. DistilBERT, from the SLM family of models and only requiring a fraction of resources to make predictions performed significantly better (over 90% in both accuracy and recall). These outcomes indicate that LLMs, particularly those targeting specific languages, are significantly superior in performance for sentiment analysis. The results seem to imply that models like these can be utilized to study the phenomenon and further enhance user engagement with startups. Future work will involve the further development of more language specific BERT versions and expanding LLMs in other ways within startup spaces.
Deepfake detection and medical image authentication via a blockchain-integrated deep learning framework Keshavmurthy Shankar Krupa; Kiran Chandrappa
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4733-4742

Abstract

The recent advances in deepfake technologies have increased the risk of adversarial attacks on clinical content. The existing schemes towards image repositories are mostly designed considering a centralized system to store and verify medical images. However, they are susceptible to tampering and data breaches. This paper presents a blockchain-integrated deep learning (DL) framework for a secure and verifiable medical image authentication process. The first stage presents Efficient-vision-transformer (ViT), a hybrid learning model designed based on the joint approach of a convolutional EfficientNet model and vision transformers for detecting deepfake images. The second stage adopts a blockchain approach where the verified real images are subjected to hash computation and then stored on decentralized storage nodes using the interplanetary file system (IPFS). The experimental results showed an impressive detection accuracy of 99% by Efficient-ViT, and it is robust against various adversarial attacks. The blockchain module further ensures tamper-proofing, transparency, and long-term data integrity.
Machine learning and failure mode and effects analysis for sustainable medical equipment spare parts management Fabian Halley Pata Alban Dattu; Syed Tramizi Syed Shazali; Shirley Johnathan Tanjong; Abdul Rani Achmed Abdullah; Nurlaila Rosli
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 5: October 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i5.pp4273-4282

Abstract

Spare parts for essential medical equipment are vital to ensure uninterrupted healthcare provision, especially in areas where resources are limited. This paper presents an integrative approach that integrates failure mode and effects analysis (FMEA) with supervised machine learning (ML) models in order to allow better predictive maintenance and optimize management of spare parts. In order to increase the accuracy of the forecast, FMEA was first used to compute risk priority numbers (RPNs) of the components, quantifying the risk of each component, and then using it as an important input feature to the ML algorithms. Data from six hospitals and ten critical medical devices, such as ventilators, defibrillators, dialysis machines, and computed tomography (CT) scanners, were analyzed to identify high-risk failure modes and prioritize maintenance tasks. Three different supervised learning models (random forest (RF), artificial neural network (ANN), and support vector machine (SVM)) was built and tested based on four metrics: accuracy, precision, recall, and F1-score. The precision of RF and ANN was 1.00 and SVM showed an excellent recall value of 0.94, indicating promising results for practical applications. The solution proposed consists of a hierarchical decision system, connecting risk assessment carried out by experts with data-driven forecasts, to create proactive and sustainable management of spare parts. This approach helps healthcare entities lessen the risk of equipment breakdowns, adhere to international healthcare standards, and increase system reliability and efficiency.

Page 207 of 215 | Total Record : 2143

 205   206   207   208   209 

Filter by Year

2012 2026


Reset
Filter By Issues
All Issue Vol 15, No 5: October 2026 Vol 15, No 4: August 2026 Vol 15, No 3: June 2026 Vol 15, No 2: April 2026 Vol 15, No 1: February 2026 Vol 14, No 6: December 2025 Vol 14, No 5: October 2025 Vol 14, No 4: August 2025 Vol 14, No 3: June 2025 Vol 14, No 2: April 2025 Vol 14, No 1: February 2025 Vol 13, No 4: December 2024 Vol 13, No 3: September 2024 Vol 13, No 2: June 2024 Vol 13, No 1: March 2024 Vol 12, No 4: December 2023 Vol 12, No 3: September 2023 Vol 12, No 2: June 2023 Vol 12, No 1: March 2023 Vol 11, No 4: December 2022 Vol 11, No 3: September 2022 Vol 11, No 2: June 2022 Vol 11, No 1: March 2022 Vol 10, No 4: December 2021 Vol 10, No 3: September 2021 Vol 10, No 2: June 2021 Vol 10, No 1: March 2021 Vol 9, No 4: December 2020 Vol 9, No 3: September 2020 Vol 9, No 2: June 2020 Vol 9, No 1: March 2020 Vol 8, No 4: December 2019 Vol 8, No 3: September 2019 Vol 8, No 2: June 2019 Vol 8, No 1: March 2019 Vol 7, No 4: December 2018 Vol 7, No 3: September 2018 Vol 7, No 2: June 2018 Vol 7, No 1: March 2018 Vol 6, No 4: December 2017 Vol 6, No 3: September 2017 Vol 6, No 2: June 2017 Vol 6, No 1: March 2017 Vol 5, No 4: December 2016 Vol 5, No 3: September 2016 Vol 5, No 2: June 2016 Vol 5, No 1: March 2016 Vol 4, No 4: December 2015 Vol 4, No 3: September 2015 Vol 4, No 2: June 2015 Vol 4, No 1: March 2015 Vol 3, No 4: December 2014 Vol 3, No 3: September 2014 Vol 3, No 2: June 2014 Vol 3, No 1: March 2014 Vol 2, No 4: December 2013 Vol 2, No 3: September 2013 Vol 2, No 2: June 2013 Vol 2, No 1: March 2013 Vol 1, No 4: December 2012 Vol 1, No 3: September 2012 Vol 1, No 2: June 2012 Vol 1, No 1: March 2012 More Issue