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
50 Documents
Search results for
, issue
"Vol 12, No 2: June 2023"
:
50 Documents
clear
<![CDATA[Dialect classification using acoustic and linguistic features in Arabic speech ]]>
<![CDATA[Mohammad Ali Humayun]]>;
<![CDATA[Hayati Yassin]]>;
<![CDATA[Pg Emeroylariffion Abas]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp739-746
<![CDATA[Speech dialects refer to linguistic and pronunciation variations in the speech of the same language. Automatic dialect classification requires considerable acoustic and linguistic differences between different dialect categories of speech. This paper proposes a classification model composed of a combination of classifiers for the Arabic dialects by utilizing both the acoustic and linguistic features of spontaneous speech. The acoustic classification comprises of an ensemble of classifiers focusing on different frequency ranges within the short-term spectral features, as well as a classifier utilizing the ‘i-vector’, whilst the linguistic classifiers use features extracted by transformer models pre-trained on large Arabic text datasets. It has been shown that the proposed fusion of multiple classifiers achieves a classification accuracy of 82.44% for the identification task of five Arabic dialects. This represents the highest accuracy reported on the dataset, despite the relative simplicity of the proposed model, and has shown its applicability and relevance for dialect identification tasks. ]]>
<![CDATA[Multi-channel of electroencephalogram signal in multivariable brain-computer interface ]]>
<![CDATA[Esmeralda Contessa Djamal]]>;
<![CDATA[Dimas Andhika Sury]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp618-626
<![CDATA[Brain-computer interface (BCI) usually uses Electroencephalogram (EEG) signals as an intermediate device to drive external devices directly from the brain. The development of BCI capabilities is carried out by involving multivariable EEG signals as movement commands. EEG signals are recorded using multi-channel, enriching information if it uses the suitable method and architecture. This research proposed a two-dimensional convolutional neural networks (CNN) method to recognize multi-channel EEG signals. The vertical dimension is the channel, while the horizontal is the signal sequence. Hence, the signal is connected with the information time series of the same channel and between channels simultaneously. BCI was arranged with multivariable signals, specifically motor imagery and emotion. Both variables have different characteristics, and the information is from different channels. Therefore, it needs multiple CNNs to recognize the two variables in the EEG signal. The experiment showed that the accuracy of multiple 2D-CNN increased to 94.62% compared to 85.44% of single 2D CNN. Multiple 2D-CNN gave accuracy from 82.04% to 94.62% more than multiple 1D-CNN. 2D-CNN makes the channel extraction perfect into vectors to maintain the signal sequence. Signal extraction is essential, so the used Wavelet filter upgraded accuracy from 73.75% to 94.62%.]]>
<![CDATA[Neural network-based pH and coagulation adjustment system in water treatment ]]>
<![CDATA[Oscar Ivan Vargas Mora]]>;
<![CDATA[Daiam Camilo Parrado Nieto]]>;
<![CDATA[Jairo David Cuero Ortega]]>;
<![CDATA[Javier Eduardo Martinez Baquero]]>;
<![CDATA[Robinson Jimenez Moreno]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp560-567
<![CDATA[This document presents a machine learning model development as a tool to improve chemical dosing procedure in ariari regional aqueduct (ARA). The supervised learning model has been addressed starting from the knowledge of data color, turbidity and pH at the water inlet to the aqueduct and the dosing results of type A aluminum sulfate and calcium oxide (lime) obtained through jar tests. The construction of the automatic learning model had a comprehensive implementation and improvement field through continuous system training, which allowed an optimal dosage of Aluminum Sulfate and Lime to generate an outlet pH less than 7.5 and outlet turbidity less than 8 nephelometric turbidity unit (NTU). Those outlet water parameters meet the ministry of social protection criteria in Colombia. Also, a virtual jar test was created to reduce the time required to obtain chemical dosing values to less than a minute. In contrast, a laboratory test takes approximately a half-hour to displays results.]]>
<![CDATA[Facial expression recognition of masked faces using deep learning ]]>
<![CDATA[Boutaina Hdioud]]>;
<![CDATA[Mohammed El Haj Tirari]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp921-930
<![CDATA[Facial expression recognition (FER) represents one of the most prevalent forms of interpersonal communication, which contains rich emotional information. But it became even more challenging during the times of COVID, where face masks became a mandatory protection measure, leading to the challenge of occluded lower-face during facial expression recognition. In this study, deep convolutional neural network (DCNN) represents the core of both our full-face FER system and our masked face FER model. The focus was on incorporating knowledge distillation in transfer learning between a teacher model, which is the full-face FER DCNN, and the student model, which is the masked face FER DCNN via the combination of both the loss from the teacher soft-labels vs the student soft labels and the loss from the dataset hard-labels vs the student hard-labels. The teacher-student architecture used FER2013 and a masked customized version of FER2013 as datasets to generate an accuracy of 69% and 61% respectively. Therefore, the study proves that the process of knowledge distillation may be used as a way for transfer learning and enhancing accuracy as a regular DCNN model (student only) would result in 46% accuracy compared to our approach (61% accuracy).]]>
<![CDATA[Modeling of an artificial intelligence based enterprise callbot with natural language processing and machine learning algorithms ]]>
<![CDATA[Imad Aattouri]]>;
<![CDATA[Hicham Mouncif]]>;
<![CDATA[Mohamed Rida]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp943-955
<![CDATA[The management of customer services by telephone encounters several problems: an uncontrollable flow of calls, complicated resource management, a very high cost of service, and more. Opportunities to improve the quality of service, save time and money triggered the widespread implementation of artificial intelligence (AI) based callbot. This article outlines the straightforward workflow developed to model the architecture of the callbot. Therefore, several algorithms were evaluated and compared based on real knowledge of a call center of an insurance society. The algorithms considered are: k-nearest neighbours (KNN), support vector machine (SVM), random forests (RF), logistic regression (LR), and Na¨ıve Bayes (NB). The comparison criteria are: correct responses, response time, accuracy, Cohen’s kappa and F1 score using n-gram (1.1) and (2.2). The results obtained show that the SVM (accuracy=70.29%) presents the best results on all the comparison criteria. The comparison between the results of the human agents and the callbot shows an improvement in several levels: the cost savings are greater than 80% on all the tests carried out, the holding time decrease to 0 seconds, and the processing time (almost a third or more). The results obtained sufficiently meet the objectives of this project.]]>
<![CDATA[Machine learning classifiers for detection of glaucoma ]]>
<![CDATA[Reshma Verma]]>;
<![CDATA[Lakshmi Shrinivasan]]>;
<![CDATA[Basvaraj Hiremath]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp806-814
<![CDATA[Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead to loss of vision. Which is known as ‘silent thief of sight’. There are several methods in which the disease can be treated, if detected at an early stage It is not possible for any technology, including artificial intelligence, to replace a doctor. However, it is possible to develop a model based on several classical image processing algorithms, combined with artificial intelligence that can detect onset of glaucoma based on certain parameters of the retinal fundus. This model would play an important role in early detection of the disease and assist the doctor. The traditional methods to detect glaucoma, as efficient as they may be, are usually expensive, a machine learning approach to diagnose from fundus images and accurately classify its severity can be considered to be efficient. Here we propose support vector machine (SVM) method to segregate, train the models using a high-end graphics processor unit (GPU) and augment the hull convex approach to boost the accuracy of the image processing mechanisms along with distinguishing the different stages of glaucoma. A web application for the screening process has also been adopted.]]>
<![CDATA[Cross-checked screening application for reliable categorisation of familial hypercholesterolaemia: design and development of the prototype ]]>
<![CDATA[Marshima Mohd Rosli]]>;
<![CDATA[Muthukkaruppan Annamalai]]>;
<![CDATA[Noor Alicezah Mohd Kasim]]>;
<![CDATA[Chua Yung-An]]>;
<![CDATA[Hapizah Mohd Nawawi]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp704-713
<![CDATA[The paper describes the development of a computer-based familial hypercholesterolemia (FH) screening application (FH CatScreen©). The application facilitates automatic scoring and categorisation of patients by medical practitioners based on four well-known FH diagnostic criteria. In the absence of a FH diagnostic criterion for Malaysian population, these four diagnostic criteria are commonly used criteria to classify patients FH severity levels to manage early interventions. We applied an adaptive software development approach comprising planning, development and validation phases to develop FH CatScreen©. A user study involving thirty medical practitioners was conducted to evaluate the effectiveness and usability of FH CatScreen©. The study showed that FH CatScreen© was able to provide a more correct, faster and better-informed assessment compared to the traditional paper-based method. The study further showed that FH CatScreen© has a good degree of performance and acceptance by the participants. The participants indicated that the simultaneous use of the four diagnostic criteria in FH CatScreen© has assisted them to compare the outcomes of each of the criterion side-by-side. It allowed them to decide on the severity of patient condition with high confidence. FH CatScreen© has demonstrated its expediency and efficacy in collecting the data on FH incidence and prevalence in Malaysia.]]>
<![CDATA[A collaborated genetic with lion optimization algorithms for improving the quality of forwarding in a vehicular ad-hoc network ]]>
<![CDATA[Sami Abduljabbar Rashid]]>;
<![CDATA[Mustafa Maad Hamdi]]>;
<![CDATA[Lukman Audah]]>;
<![CDATA[Mohammed Ahmed Jubair]]>;
<![CDATA[Mustafa Hamid Hassan]]>;
<![CDATA[Mohammed Salah Abood]]>;
<![CDATA[Salama A. Mostafa]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp667-677
<![CDATA[Vehicular ad-hoc network (VANET) is dynamic and it works on various noteworthy applications in intelligent transportation systems (ITS). In general, routing overhead is more in the VANETs due to their properties. Hence, need to handle this issue to improve the performance of the VANETs. Also due to its dynamic nature collision occurs. Up till now, we have had immense complexity in developing the multi-constrained network with high quality of forwarding (QoF). To solve the difficulties especially to control the congestion this paper introduces an enhanced genetic algorithmbased lion optimization for QoF-based routing protocol (EGA-LOQRP) in the VANET network. Lion optimization routing protocol (LORP) is an optimization-based routing protocol that can able to control the network with a huge number of vehicles. An enhanced genetic algorithm (EGA) is employed here to find the best possible path for data transmission which leads to meeting the QoF. This will result in low packet loss, delay, and energy consumption of the network. The exhaustive simulation tests demonstrate that the EGA-LOQRP routing protocol improves performance effectively in the face of congestion and QoS assaults compared to the previous routing protocols like Ad hoc on-demand distance vector (AODV), ant colony optimization-AODV (ACO-AODV) and traffic aware segmentAODV (TAS-AODV).]]>
<![CDATA[Iban plaited mat motif classification with adaptive smoothing ]]>
<![CDATA[Silvia Joseph]]>;
<![CDATA[Irwandi Hipiny]]>;
<![CDATA[Hamimah Ujir]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp840-850
<![CDATA[Decorative mats plaited by the Iban communities in Borneo contains motifs that reflect their traditional beliefs. Each motif has its own special meaning and taboos. A typical mat motif contains multiple smaller patterns that surround the main motif hence is likely to cause misclassification. We introduce a classification framework with adaptive sampling to remove smaller features whilst retaining larger (and discriminative) image structures. Canny filter and probabilistic hough transform are gradually applied to a clean greyscale image until a threshold value pertaining to the image’s structural information is reached. Morphological dilation is then applied to improve the appearance of the retained edges. The resulting image is described using binary robust invariant scalable keypoints (BRISK) features with random sample consensus (RANSAC). We reported the classification accuracy against six common image deformations at incremental degrees: scale+rotation, viewpoint, image blur, joint photographic experts group (JPEG) compression, scale and illumination. From our sensitivity analysis, we found the optimal threshold for adaptive smoothing to be 75.0%. The optimal scheme obtained 100.0% accuracy for JPEG compression, illumination, and viewpoint set. Using adaptive smoothing, we achieved an average increase in accuracy of 11.0% compared to the baseline.]]>
<![CDATA[Information system based on multi-value classification of fully connected neural network for construction management ]]>
<![CDATA[Tetyana Honcharenko]]>;
<![CDATA[Roman Akselrod]]>;
<![CDATA[Andrii Shpakov]]>;
<![CDATA[Oleksandr Khomenko]]>
<![CDATA[ IAES International Journal of Artificial Intelligence (IJ-AI) Vol 12, No 2: June 2023 ]]>
Publisher : <![CDATA[Institute of Advanced Engineering and Science ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijai.v12.i2.pp593-601
<![CDATA[This study is devoted to solving the problem to determine the professional adaptive capabilities of construction management staff using artificial intelligence systems. It is proposed fully connected feed-forward neural network (FCF-FNN) architecture and performed empirical modeling to create a data set. Model of artificial intelligence system allows evaluating the processes in an FCF-FNN during the execution of multi-value classification of professional areas. A method has been developed for the training process of a machine learning model, which reflects the internal connections between the components of an artificial intelligence system that allow it to “learn” from training data. To train the neural network, a data set of 35 input parameters and 29 output parameters was used; the amount of data in the set is 936 data lines. Neural network training occurred in the proportion of 10% and 90%, respectively. Results of this study research can be used to further improve the knowledge and skills necessary for successful professional realization.]]>
