IAES International Journal of Artificial Intelligence (IJ-AI)
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.
Articles
81 Documents
Search results for
, issue
"Vol 14, No 3: June 2025"
:
81 Documents
clear
<![CDATA[Assured time series forecasting using inertial measurement unit, neural networks, and state estimators ]]>
<![CDATA[Kulkarni, Ashvini]]>;
<![CDATA[Beulet Paul, Augusta Sophy]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp1870-1883
<![CDATA[Pedestrian dead reckoning (PDR) technology has be come an important method for predicting the position of an object or person. Sensor-based positioning is widely used because of its readily available hardware and acceptable accuracy, especially with PDR algorithms integrated with machine learning and deep learning. There are two challenges in this context. Conventional state-estimator methods suffers from dynamics, making the deployment and management of nonlinear dynamics become difficult. Training an effective neural network model with a few inertial measurement unit (IMU) samples is also challenging. This study investigates the integration and comparison of advanced state estimation algorithms such as the Kalman filter (KF), extended Kalman filter (EKF), and sigma point Kalman filter (SPKF) with deep neural networks, including multi-layer perceptron (MLP), convolutional neural network (CNN), and long short-term memory (LSTM). The aim is to improve the reliability of forecasting and prediction tasks, particularly when processing IMU data. This study conducts a comprehensive performance comparison between state estimators integration with deep learning models, evaluating their effectiveness in addressing the challenges of estimation and prediction. The preliminary results show that the feature forecasting rate of the proposed method can reach a root mean square error (RMSE) value of 0.31 (EKF-LSTM) and 1.50 (SPKF-LSTM).]]>
<![CDATA[The main weaknesses of using Manhattan distance for solving sliding tile puzzles ]]>
<![CDATA[Nayef Al Refai, Mohammed]]>;
<![CDATA[Mohammed Jamhawi, Zeyad]]>;
<![CDATA[Ali Otoom, Ahmed]]>;
<![CDATA[Al-Momani, Adai]]>;
<![CDATA[Khafajeh, Hayel]]>;
<![CDATA[Atoum, Issa]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp2423-2432
<![CDATA[Heuristics are a big improvement over blind searching in pathfinding. The node's test, run, and finish time are reasonable. Artificial intelligence (AI) uses Manhattan distance (MD), a good and simple heuristic, in various subjects to reduce the number of exploring nodes while requiring fewer calculations. The MD heuristics examined approximately 25 times fewer states than the blind search. Unfortunately, can’t reach the goal of pathfinding when the domain size increases, as it becomes similar to brute force or blind search algorithm results. Previous studies have concentrated on MD's weakness, specifically its low bound value for calculation results, and attempted to improve this value in various ways. Unfortunately, to our knowledge, none of the presented research has been able to find the optimal path for all slide tile puzzle sizes. This work discusses the detailed reasons for the low bound value and other related factors that contribute to its weakness. This paper discovered that the distribution of MD values within the domain, not lowbound values, is the critical issue that complicates the search. The MD's summation method for all tiles has an impact on the calculated duplication values. The total number of nodes in the optimal path also affects the search performance.]]>
<![CDATA[Review on class imbalance techniques to strengthen model prediction ]]>
<![CDATA[Putta, Hemalatha]]>;
<![CDATA[Amalanathan, Geetha Mary]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp1727-1742
<![CDATA[Data is a fundamental component in various fields, including science, business, health care, and technology. It is often processed, stored, and analyzed using computer systems and software applications. The importance of data lies in its ability to provide valuable insights, drive innovation, and improve decision-making processes. However, it’s essential to handle and manage data responsibly to address privacy and ethical considerations. Data mining (DM) involves discovering patterns, trends, correlations, or useful information from large datasets. Data dredging or DM and machine learning (ML) are closely related fields that both involve the analysis of data to discover patterns and make predictions. DM focuses on extracting knowledge from data; ML emphasizes the development of algorithms that can do analysis. The two fields are interconnected, and the techniques from one state of art integrated into the processes of the other. In ML the class imbalance problem occurs due to the class distribution in the training data is not equal. Imbalanced classification refers to a condition where a particular class (minority class) is under represented parallelled to another class (majority class) in a dataset. This paper furthermore emphasizes on the synthetic minority oversampling technique (SMOTE) variants employed by the researchers, and highlights the limitations the work.]]>
<![CDATA[Comparison of deep learning models: CNN and VGG-16 in identifying pornographic content ]]>
<![CDATA[Chandra, Reza]]>;
<![CDATA[Suhendra, Adang]]>;
<![CDATA[Yuniar Banowosari, Lintang]]>;
<![CDATA[Prihandoko, Prihandoko]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp1884-1899
<![CDATA[In 2020, a total of 59,741 websites were blocked by the Indonesian government due to containing negative content, including pornography, with 14,266 websites falling into this category. However, these blocked websites could still be accessed by the public using virtual private networks (VPNs). This prompted the research idea to quickly identify pornographic content. This study aims to develop a system capable of identifying websites suspected of containing pornographic image content, using a deep learning approach with convolutional neural network (CNN) and visual geometry group 16 (VGG-16) model. The two models were then explored comprehensively and holistically to determine which model was most effective in detecting pornographic content quickly. Based on the findings of the comparison between testing the CNN and VGG-16 models, research results showed that the best test results were obtained in the eighth experiment using the CNN model at an epoch value level of 50 and a learning rate of 0.001 of 0.9487 or 94.87%. This can be interpreted that the CNN model is more effective in detecting pornographic content quickly and accurately compared to using the VGG-16 model.]]>
<![CDATA[Co-training pseudo-labeling for text classification with support vector machine and long short-term memory ]]>
<![CDATA[Handayani, Sri]]>;
<![CDATA[Isnanto, Rizal]]>;
<![CDATA[Warsito, Budi]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp2158-2168
<![CDATA[The scarcity of labeled data may hamper training text-processing models. In response to this issue, a novel and intriguing strategy that combines the co-training method and pseudo-labeling design is applied to enhance the model's performance. This method, a component of an efficient semi-supervised learning paradigm for processing and comprehending text, is a fresh perspective in the field. The model, which combines a support vector machine (SVM) for classification and long short-term memory (LSTM) for text sequence interpretation, is a unique approach. By introducing samples that may be marginalized in the labeled data, the co-training approach could help solve the class imbalance problem by using a small amount of labeled data and the rest unlabeled. This study assesses the model's performance using a student dataset from higher education institutions to establish a threshold for each model's degree of confidence and ascertain how much the model can be generalized depending on the threshold. The SVM threshold was calculated as >=0.88, and the LSTM threshold was calculated as >=0.5 using a mixture of confidence metrics.]]>
<![CDATA[Advancements in abstractive text summarization: a deep learning approach ]]>
<![CDATA[Suliman, Wesam]]>;
<![CDATA[Yaseen, Amer]]>;
<![CDATA[Hamada, Nuha]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp2315-2327
<![CDATA[With the rapid growth of data, text summarization has become vital for extracting key information efficiently. While extractive text summarization models are widely available, they often produce redundant outputs with limited capability of generating human-like summaries. Abstractive summarization, which generates new phrases and rephrases content, remains underexplored due to its complexity. This paper addresses this gap by developing an abstractive deep learning model using an encoder-decoder architecture supported with an attention mechanism. Trained on the dataset of Amazon Food Reviews, the model generates contextually rich and semantically accurate summaries. The model’s evaluation using BLEU and ROUGE metrics demonstrated promising results, with a score of 0.641 for BLEU, 0.520 for ROUGE-1, 0.345 for ROUGE-2, 0.461 for ROUGE-L and 0.428 for ROUGE-W, indicating coherence and structural integrity. This research highlights the potential of deep learning in addressing the limitations of classical methods and suggests opportunities for future advancements, such as scaling the model with larger datasets and integrating transformer-based techniques for improved summarization across diverse applications.]]>
<![CDATA[Hybrid semantic model based on machine learning for sentiment classification of consumer reviews ]]>
<![CDATA[Rajidurai Parvathy, Palaniraj]]>;
<![CDATA[Mohankumar, Nagarajan]]>;
<![CDATA[Shobiga, Rajendran]]>;
<![CDATA[Mitra Thakur, Gour Sundar]]>;
<![CDATA[Bandaru, Mamatha]]>;
<![CDATA[Sujatha, Velusamy]]>;
<![CDATA[Sujatha, Shanmugam]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp2001-2011
<![CDATA[Digital information is regularly produced from a variety of sources, including social media and customer service reviews. For the purpose of increasing customer happiness, this written data must be processed to extract user comments. Consumers typically share comments and thoughts about consumable items, technological goods, and services supplied for payment in the modern period of consumerism with simple access to social networking globe. Each object has a plethora of remarks or thoughts that demand special attention due to their sentimental worth, especially in the written portions. The goal of the current project is to do sentiment prediction on the Amazon Electronics, Kindle, and Gift Card datasets. In order to predict sentiment and evaluate utilizing many executions evaluates admitting accuracy, recall, and F1-score, a hybrid soft voting ensemble method that combines lexical and ensemble methodologies is proposed in this study. In addition to calculating a subjectivity score and sentiment score, this study also suggests a non-interpretive sentiment class label that may be used to assess the sign of the evaluations applying suggested method for sentiment categorization. The effectiveness of our suggested ensemble model is examined using datasets from Amazon customer product reviews, and we found an improvement of 2-5% in accuracy compared to the current state-of-the-art ensemble method.]]>
<![CDATA[An optimized transfer learning-based approach for Crocidolomia pavonana larvae classification ]]>
<![CDATA[Risnawati, Risnawati]]>;
<![CDATA[Rodiah, Rodiah]]>;
<![CDATA[Madenda, Sarifuddin]]>;
<![CDATA[Tri Susetianingtias, Diana]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp2270-2281
<![CDATA[The increasing demand for mustard greens has driven farmers to continuously improve mustard greens cultivation. One of the challenges in mustard greens cultivation is the presence of insect pests. A significant pest in mustard greens is Crocidolomia pavonana (C. pavonana). C. pavonana damages plants by feeding on various parts, especially the leaves. The initial step in controlling them is insect pest monitoring. Monitoring aims to establish the control threshold. C. pavonana larvae have four instar stages: instar 1, 2, 3, and 4. Identification of the instar larval stages utilizes deep convolutional neural network (CNN) to classify C. Pavonana larvae on mustard greens using ResNet50V2 and DenseNet169 architectures optimized to enhance classification accuracy. The classification evaluation results show that both DenseNet169 and ResNet50V2 models achieve high accuracy, with DenseNet169 reaching the highest accuracy at 97.1%, while ResNet50V2 achieves an accuracy of 94.2%. The lower loss values on the test data compared to the validation data indicate that the deep learning models have successfully captured the patterns in C. pavonana images for classification. This classification process is expected to be one of the activities in monitoring the instar larvae to improve the accuracy of insecticide spraying and enhance mustard greens production.]]>
<![CDATA[Improvisation in detection of pomegranate leaf disease using transfer learning techniques ]]>
<![CDATA[M. Metagar, Shivappa]]>;
<![CDATA[A. Walikar, Gyanappa]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp1930-1939
<![CDATA[To provide the growing world population with food and satisfy their fundamental requirements, agriculture is a vital industry. The cultivation of cereals and vegetables is indispensable for both human sustenance and the worldwide economy. Many farmers in rural areas suffer substantial losses because they rely on manual monitoring of crops and lack sufficient information and disease detection methods. Digital farming techniques may provide a novel way to swiftly and simply identify illnesses in the leaves of plants. This article uses image processing and transfer learning techniques for identifying plant leaf ailments and taking preventative action in the agriculture business in order to address these problems. Global food security and agricultural productivity are seriously threatened by leaf disease. Crop losses may be considerably decreased, and crop output can be increased by promptly identifying and diagnosing leaf diseases. Deep learning can mitigate the adverse impact of artificially picking disease spot data, enhance objectivity in extracting plant disease traits, and expedite the advancement of new technologies. This article presents a novel approach using deep learning to diagnose leaf diseases. This article advances the development of efficient and successful techniques for recognizing and diagnosing leaf diseases, which will eventually aid farmers and maintain the security of the world's food supply.]]>
<![CDATA[Novel preemptive intelligent artificial intelligence-model for detecting inconsistency during software testing ]]>
<![CDATA[Govinda, Sangeetha]]>;
<![CDATA[Prasanthi, B. G.]]>;
<![CDATA[Vincent, Agnes Nalini]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 3: June 2025 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v14.i3.pp1781-1789
<![CDATA[The contribution of artificial intelligence (AI)-based modelling is highly significant in automating the software testing process; thereby enhancing the cost, resources, and productivity while performing testing. Review of existing AI-models towards software testing showcases yet an open-scope for further improvement as yet the conventional AI-model suffers from various challenges especially in perspective of test case generation. Therefore, the proposed scheme presents a novel preemptive intelligent computational framework that harnesses a unique ensembled AI-model for generating and executing highly precise and optimized test-cases resulting in an outcome of adversary or inconsistencies associated with test cases. The ensembled AI-model uses both unsupervised and supervised learning approaches on publicly available outlier dataset. The benchmarked outcome exhibits supervised learning-based AI-model to offer 21% of reduced error and 1.6% of reduced processing time in contrast to unsupervised scheme while performing software testing.]]>
