Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp1794-1804
Tedy Setiadi, Mohd Ridzwan Yaakub, Azuraliza Abu Bakar
Community detection is a field of interest in social networks. Many new methods have emerged for community detection solution, however the modularity optimization method is the most prominent. Community detection based on modularity optimization (CDMO) has fundamental problems in the form of solution degeneration and resolution limits. From the two problems, the resolution limit is more concerned because it affects the resulting community's quality. During the last decade, many studies have attempted to address the problems, but so far they have been carried out partially, no one has thoroughly discussed efforts to improve the quality of CDMO. In this paper, we aim to investigate works in handling resolution limit and improving the quality of CDMO, along with their strengths and limitations. We derive six categories of strategies to improve the quality of CDMO, namely developing multi-resolution modularity, creating local modularity, creating modularity density, creating new metrics as an alternative to modularity, creating new quality metrics as a substitute for modularity, involving node attributes in determining community detection, and extending the single objective function into a multi-objective function. These strategies can be used as a guide in developing community detection methods. By considering network size, network type, and community distribution, we can choose the appropriate strategy in improving the quality of community detection.
{"title":"Strategies for improving the quality of community detection based on modularity optimization","authors":"Tedy Setiadi, Mohd Ridzwan Yaakub, Azuraliza Abu Bakar","doi":"10.11591/ijai.v13.i2.pp1794-1804","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp1794-1804","url":null,"abstract":"Community detection is a field of interest in social networks. Many new methods have emerged for community detection solution, however the modularity optimization method is the most prominent. Community detection based on modularity optimization (CDMO) has fundamental problems in the form of solution degeneration and resolution limits. From the two problems, the resolution limit is more concerned because it affects the resulting community's quality. During the last decade, many studies have attempted to address the problems, but so far they have been carried out partially, no one has thoroughly discussed efforts to improve the quality of CDMO. In this paper, we aim to investigate works in handling resolution limit and improving the quality of CDMO, along with their strengths and limitations. We derive six categories of strategies to improve the quality of CDMO, namely developing multi-resolution modularity, creating local modularity, creating modularity density, creating new metrics as an alternative to modularity, creating new quality metrics as a substitute for modularity, involving node attributes in determining community detection, and extending the single objective function into a multi-objective function. These strategies can be used as a guide in developing community detection methods. By considering network size, network type, and community distribution, we can choose the appropriate strategy in improving the quality of community detection.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"13 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141229699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp1845-1857
Ashraf Saad Shewail, Hala H. Zayed, Neven A. M. Elsayed
In recent times, there has been an increase in the stability and integration of augmented reality (AR) technology in everyday applications. AR relies on tracking techniques to capture the characteristics of the surrounding environment. Tracking falls into two categories: outdoor and indoor. While outdoor tracking predominantly relies on the global positioning system (GPS), it is performance indoors is hindered by imprecise GPS signals. Indoor tracking offers a solution for navigating complex indoor environments. This paper introduces an indoor tracking system that combines smartphone sensor data and computer vision using the oriented features from accelerated and segments test and rotated binary robust independent elementary features (ORB) algorithm for feature extraction, along with brute force match (BFM) and k-nearest neighbor (KNN) for matching. This approach outperforms previous systems, offering efficient navigation without relying on pre-existing maps. The system uses the A* algorithm to find the shortest path and cloud computing for data storage. Experimental results demonstrate an impressive 99% average accuracy within a 7-10 cm error range, even in scenarios with varying distances. Moreover, all users successfully reached their destinations during the experiments. This innovative model presents a promising advancement in indoor tracking, enhancing the accuracy and effectiveness of navigation in complex indoor spaces
{"title":"Real-time indoor tracking for augmented reality using computer vision technique","authors":"Ashraf Saad Shewail, Hala H. Zayed, Neven A. M. Elsayed","doi":"10.11591/ijai.v13.i2.pp1845-1857","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp1845-1857","url":null,"abstract":"In recent times, there has been an increase in the stability and integration of augmented reality (AR) technology in everyday applications. AR relies on tracking techniques to capture the characteristics of the surrounding environment. Tracking falls into two categories: outdoor and indoor. While outdoor tracking predominantly relies on the global positioning system (GPS), it is performance indoors is hindered by imprecise GPS signals. Indoor tracking offers a solution for navigating complex indoor environments. This paper introduces an indoor tracking system that combines smartphone sensor data and computer vision using the oriented features from accelerated and segments test and rotated binary robust independent elementary features (ORB) algorithm for feature extraction, along with brute force match (BFM) and k-nearest neighbor (KNN) for matching. This approach outperforms previous systems, offering efficient navigation without relying on pre-existing maps. The system uses the A* algorithm to find the shortest path and cloud computing for data storage. Experimental results demonstrate an impressive 99% average accuracy within a 7-10 cm error range, even in scenarios with varying distances. Moreover, all users successfully reached their destinations during the experiments. This innovative model presents a promising advancement in indoor tracking, enhancing the accuracy and effectiveness of navigation in complex indoor spaces","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"10 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141229764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp1625-1631
Bahtiar Imran, Lalu Delsi Samsumar, Ahmad Subki, Zaeniah Zaeniah, Salman Salman, Muhammad Rijal Alfian
This research uses the gray level co-occurrence matrix (GLCM) and artificial neural networks to classify COVID-19 images based on chest X-ray images. According to previous studies, there has never been a researcher who has integrated GLCM with artificial neural networks. Epochs 10, 30, 50, 70, 100, and 120 were used in this research. The total number of data points used in this investigation was 600, divided into 300 normal chests and 300 COVID-19 data points. Epoch 10 had 91% accuracy, epoch 30 had 91% accuracy, epoch 50 had 92% accuracy, epoch 70 had 91% accuracy, epoch 100 had 92% accuracy, and epoch 120 had 90% accuracy in categorization. As indicated by the results of the classification tests, combining GLCM and artificial neural networks can produce good results; a combination of these methods can yield a classification for COVID-19.
{"title":"Combination of gray level co-occurrence matrix and artificial neural networks for classification of COVID-19 based on chest X-ray images","authors":"Bahtiar Imran, Lalu Delsi Samsumar, Ahmad Subki, Zaeniah Zaeniah, Salman Salman, Muhammad Rijal Alfian","doi":"10.11591/ijai.v13.i2.pp1625-1631","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp1625-1631","url":null,"abstract":"This research uses the gray level co-occurrence matrix (GLCM) and artificial neural networks to classify COVID-19 images based on chest X-ray images. According to previous studies, there has never been a researcher who has integrated GLCM with artificial neural networks. Epochs 10, 30, 50, 70, 100, and 120 were used in this research. The total number of data points used in this investigation was 600, divided into 300 normal chests and 300 COVID-19 data points. Epoch 10 had 91% accuracy, epoch 30 had 91% accuracy, epoch 50 had 92% accuracy, epoch 70 had 91% accuracy, epoch 100 had 92% accuracy, and epoch 120 had 90% accuracy in categorization. As indicated by the results of the classification tests, combining GLCM and artificial neural networks can produce good results; a combination of these methods can yield a classification for COVID-19.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"38 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141233077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2275-2282
Mosleh M. Abualhaj, A. Abu-Shareha, Qusai Y. Shambour, S. Al-Khatib, Mohammad O. Hiari
Malicious software, also referred to as malware, poses a serious threat to computer networks, user privacy, and user systems. Effective cybersecurity depends on the correct detection and classification of malware. In order to improve its effectiveness, the K-Nearest Neighbors (KNN) method is applied systematically in this study to the task of malware detection. The study investigates the effect of the number of neighbors (K) parameter on the KNN's performance. MalMem-2022 malware datasets and relevant evaluation criteria like accuracy, precision, recall, and F1-score will be used to assess the efficacy of the suggested technique. The experiments evaluate how parameter tuning affects the accuracy of malware detection by comparing the performance of various parameter setups. The study findings show that careful parameter adjustment considerably boosts the KNN method's malware detection capability. The research also highlights the potential of KNN with parameter adjustment as a useful tool for malware detection in real-world circumstances, allowing for prompt and precise identification of malware.
{"title":"Tuning the k value in k-nearest neighbors for malware detection","authors":"Mosleh M. Abualhaj, A. Abu-Shareha, Qusai Y. Shambour, S. Al-Khatib, Mohammad O. Hiari","doi":"10.11591/ijai.v13.i2.pp2275-2282","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2275-2282","url":null,"abstract":"Malicious software, also referred to as malware, poses a serious threat to computer networks, user privacy, and user systems. Effective cybersecurity depends on the correct detection and classification of malware. In order to improve its effectiveness, the K-Nearest Neighbors (KNN) method is applied systematically in this study to the task of malware detection. The study investigates the effect of the number of neighbors (K) parameter on the KNN's performance. MalMem-2022 malware datasets and relevant evaluation criteria like accuracy, precision, recall, and F1-score will be used to assess the efficacy of the suggested technique. The experiments evaluate how parameter tuning affects the accuracy of malware detection by comparing the performance of various parameter setups. The study findings show that careful parameter adjustment considerably boosts the KNN method's malware detection capability. The research also highlights the potential of KNN with parameter adjustment as a useful tool for malware detection in real-world circumstances, allowing for prompt and precise identification of malware.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"10 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141235173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp1954-1968
Meeradevi Meeradevi, S. B. J., Swetha B. N.
In the realm of natural language processing (NLP), a diverse array of language models has emerged, catering to a wide spectrum of tasks, ranging from speaker recognition and auto-correction to sentiment analysis and stock prediction. The significance of language models in enabling the execution of these NLP tasks cannot be overstated. This study proposes an approach to enhance accuracy by leveraging a hybrid language model, combining the strengths of long short-term memory (LSTM) and gated recurrent unit (GRU). LSTM excels in preserving long-term dependencies in data, while GRU's simpler gating mechanism expedites the training process. The research endeavors to evaluate four variations of this hybrid model: LSTM, GRU, bidirectional long short-term memory (Bi-LSTM), and a combination of LSTM with GRU. These models are subjected to rigorous testing on two distinct datasets: one focused on IBM stock price prediction, and the other on Jigsaw toxic comment classification (sentiment analysis). This work represents a significant stride towards democratizing NLP capabilities, ensuring that even in resource-constrained settings, NLP models can exhibit improved performance. The anticipated implications of these findings span a wide spectrum of real-world applications and hold the potential to stimulate further research in the field of NLP.
{"title":"Evaluating the machine learning models based on natural language processing tasks","authors":"Meeradevi Meeradevi, S. B. J., Swetha B. N.","doi":"10.11591/ijai.v13.i2.pp1954-1968","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp1954-1968","url":null,"abstract":"In the realm of natural language processing (NLP), a diverse array of language models has emerged, catering to a wide spectrum of tasks, ranging from speaker recognition and auto-correction to sentiment analysis and stock prediction. The significance of language models in enabling the execution of these NLP tasks cannot be overstated. This study proposes an approach to enhance accuracy by leveraging a hybrid language model, combining the strengths of long short-term memory (LSTM) and gated recurrent unit (GRU). LSTM excels in preserving long-term dependencies in data, while GRU's simpler gating mechanism expedites the training process. The research endeavors to evaluate four variations of this hybrid model: LSTM, GRU, bidirectional long short-term memory (Bi-LSTM), and a combination of LSTM with GRU. These models are subjected to rigorous testing on two distinct datasets: one focused on IBM stock price prediction, and the other on Jigsaw toxic comment classification (sentiment analysis). This work represents a significant stride towards democratizing NLP capabilities, ensuring that even in resource-constrained settings, NLP models can exhibit improved performance. The anticipated implications of these findings span a wide spectrum of real-world applications and hold the potential to stimulate further research in the field of NLP. ","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"28 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141232860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2081-2092
Samah W. G. AbuSalim, Nordin Zakaria, Aarish Maqsood, Abdul Saboor, Yew Kwang Hooi, Norehan Mokhtar, Said Jadid Abdulkadir
Accurate detection and classification of teeth is the first step in dental disease diagnosis. However, the same class of tooth exhibits significant variations in surface appearance. Moreover, the complex geometrical structure poses challenges in learning discriminative features among the different tooth classes. Due to these complex features, tooth classification is one of the challenging research domains in deep learning. To address the aforementioned issues, the presented study proposes discriminative local feature extraction at different granular levels using YOLO models. However, this necessitates a granular intra-oral image dataset. To facilitate this requirement, a dataset at three granular levels (two, four, and seven teeth classes) is developed. YOLOv5, YOLOv6, and YOLOv7 models were trained using 2,790 images. The results indicate superior performance of YOLOv6 for two-class classification problems. The model generated a mean average precision (mAP) value of 94%. However, as the granularity level is increased, the performance of YOLO models decreases. For, four and seven-class classification problems, the highest mAP value of 87% and 79% was achieved by YOLOv5 respectively. The results indicate that different levels of granularity play an important role in tooth detection and classification. The YOLO’s performance gradually decreased as the granularity decreased especially at the finest granular level.
{"title":"Multi-granularity tooth analysis via YOLO-based object detection models for effective tooth detection and classification","authors":"Samah W. G. AbuSalim, Nordin Zakaria, Aarish Maqsood, Abdul Saboor, Yew Kwang Hooi, Norehan Mokhtar, Said Jadid Abdulkadir","doi":"10.11591/ijai.v13.i2.pp2081-2092","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2081-2092","url":null,"abstract":"Accurate detection and classification of teeth is the first step in dental disease diagnosis. However, the same class of tooth exhibits significant variations in surface appearance. Moreover, the complex geometrical structure poses challenges in learning discriminative features among the different tooth classes. Due to these complex features, tooth classification is one of the challenging research domains in deep learning. To address the aforementioned issues, the presented study proposes discriminative local feature extraction at different granular levels using YOLO models. However, this necessitates a granular intra-oral image dataset. To facilitate this requirement, a dataset at three granular levels (two, four, and seven teeth classes) is developed. YOLOv5, YOLOv6, and YOLOv7 models were trained using 2,790 images. The results indicate superior performance of YOLOv6 for two-class classification problems. The model generated a mean average precision (mAP) value of 94%. However, as the granularity level is increased, the performance of YOLO models decreases. For, four and seven-class classification problems, the highest mAP value of 87% and 79% was achieved by YOLOv5 respectively. The results indicate that different levels of granularity play an important role in tooth detection and classification. The YOLO’s performance gradually decreased as the granularity decreased especially at the finest granular level.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"3 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141229835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2395-2402
Avinash Nagaraja, S. Sinha, Shivamurthaiah Mallaiah
For many researchers working on feature selection techniques, finding an appropriate feature from the microarray dataset has turned into a bottleneck. Researchers often create feature selection approaches and algorithms with the goal of improving accuracy in microarray datasets. The main goal of this study is to present a variety of contemporary feature selection techniques, such as Filter, Wrapper, and Embedded methods proposed for microarray datasets to work on multi-class classification problems and different approaches to enhance the performance of learning algorithms, to address the imbalance issue in the data set, and to support research efforts on microarray dataset. This study is based on Critical Review Questions (CRQ) constructed using feature election methods described in the review methodology and applied to a microarray dataset. We discussed the analysed findings and future prospects of feature selection strategies for multi-class classification issues using microarray datasets, as well as prospective ways to speed up computing environment
{"title":"Feature selection techniques for microarray dataset: a review","authors":"Avinash Nagaraja, S. Sinha, Shivamurthaiah Mallaiah","doi":"10.11591/ijai.v13.i2.pp2395-2402","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2395-2402","url":null,"abstract":"For many researchers working on feature selection techniques, finding an appropriate feature from the microarray dataset has turned into a bottleneck. Researchers often create feature selection approaches and algorithms with the goal of improving accuracy in microarray datasets. The main goal of this study is to present a variety of contemporary feature selection techniques, such as Filter, Wrapper, and Embedded methods proposed for microarray datasets to work on multi-class classification problems and different approaches to enhance the performance of learning algorithms, to address the imbalance issue in the data set, and to support research efforts on microarray dataset. This study is based on Critical Review Questions (CRQ) constructed using feature election methods described in the review methodology and applied to a microarray dataset. We discussed the analysed findings and future prospects of feature selection strategies for multi-class classification issues using microarray datasets, as well as prospective ways to speed up computing environment","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"3 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141229041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2036-2048
F. E. Nastiti, Shahrulniza Musa, Eiad Yafi
In this study, a novel approach geared toward predicting the estimated number of coronavirus disease (COVID-19) cases was developed. Combining long short-term memory (LSTM) neural networks with particle swarm optimization (PSO) along with grey wolf optimization (GWO) employ hybrid optimization algorithm techniques. This investigation utilizes COVID-19 original data from the Ministry of Health of Indonesia, period 2020-2021. The developed LSTM-PSO-GWO hybrid optimization algorithm can improve the performance and accuracy of predicting the spread of the COVID-19 virus in Indonesia. In initiating LSTM initial weights with weaknesses, using the hybrid optimization algorithm helps overcome these problems and improve model performance. The results of this study suggest that the LSTM-PSO-GWO model can be utilized as an effective and reliable predictive tool to gauge the COVID-19 virus’s spread in Indonesia.
{"title":"A hybrid deep learning optimization for predicting the spread of a new emerging infectious disease","authors":"F. E. Nastiti, Shahrulniza Musa, Eiad Yafi","doi":"10.11591/ijai.v13.i2.pp2036-2048","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2036-2048","url":null,"abstract":"In this study, a novel approach geared toward predicting the estimated number of coronavirus disease (COVID-19) cases was developed. Combining long short-term memory (LSTM) neural networks with particle swarm optimization (PSO) along with grey wolf optimization (GWO) employ hybrid optimization algorithm techniques. This investigation utilizes COVID-19 original data from the Ministry of Health of Indonesia, period 2020-2021. The developed LSTM-PSO-GWO hybrid optimization algorithm can improve the performance and accuracy of predicting the spread of the COVID-19 virus in Indonesia. In initiating LSTM initial weights with weaknesses, using the hybrid optimization algorithm helps overcome these problems and improve model performance. The results of this study suggest that the LSTM-PSO-GWO model can be utilized as an effective and reliable predictive tool to gauge the COVID-19 virus’s spread in Indonesia. ","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"10 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141230497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2155-2164
A. Muklason, Yusnardo Tendio, Helena Angelita Depari, Muhammad Arif Nuriman, Gusti Agung Premananda
In general, uncapacitated exam timetabling is conducted manually, which can be time-consuming. Many studies aim to automate and optimize uncapacitated exam timetabling. However, pinpointing the most efficient algorithm is challenging since most studies assert that their algorithms surpass previous ones. To identify the optimal algorithm, this research evaluates the performance of four algorithms: Hill climbing (HC), simulated annealing (SA), great deluge (GD), and tabu search (TS) in addressing the exam timetabling problem. The Kempe chain operator’s influence on optimization solutions is also examined. A simple random method is employed to select the low-level heuristic (LLH). The Carter (Toronto) dataset served as the test material, with each algorithm undergoing 200,000 iterations for comparison. The results indicate that the TS algorithm is superior, providing the best solution in 13 instances. The use of a tabu list enhanced the search process’s efficiency by preventing redundant modifications. The Kempe chain LLH exhibited a tendency towards achieving better solutions.
{"title":"The performance analysis of hyper-heuristics algorithms over examination timetabling problems","authors":"A. Muklason, Yusnardo Tendio, Helena Angelita Depari, Muhammad Arif Nuriman, Gusti Agung Premananda","doi":"10.11591/ijai.v13.i2.pp2155-2164","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2155-2164","url":null,"abstract":"In general, uncapacitated exam timetabling is conducted manually, which can be time-consuming. Many studies aim to automate and optimize uncapacitated exam timetabling. However, pinpointing the most efficient algorithm is challenging since most studies assert that their algorithms surpass previous ones. To identify the optimal algorithm, this research evaluates the performance of four algorithms: Hill climbing (HC), simulated annealing (SA), great deluge (GD), and tabu search (TS) in addressing the exam timetabling problem. The Kempe chain operator’s influence on optimization solutions is also examined. A simple random method is employed to select the low-level heuristic (LLH). The Carter (Toronto) dataset served as the test material, with each algorithm undergoing 200,000 iterations for comparison. The results indicate that the TS algorithm is superior, providing the best solution in 13 instances. The use of a tabu list enhanced the search process’s efficiency by preventing redundant modifications. The Kempe chain LLH exhibited a tendency towards achieving better solutions.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"85 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141231003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-01DOI: 10.11591/ijai.v13.i2.pp2165-2172
Jishnu Kaitholikkal Sasi, Arthi Balakrishnan
Phishing attacks pose a constant threat to online security, necessitating the development of efficient tools for identifying malicious URLs. In this article, we propose a novel approach to detect phishing URLs employing a generative adversarial network (GAN) with a variational autoencoder (VAE) as the generator and a transformer model with self-attention as the discriminator. The VAE generator is trained to produce synthetic URLs. In contrast, the transformer discriminator uses its self-attention mechanism to focus on the different parts of the input URLs to extract crucial features. Our model uses adversarial training to distinguish between legitimate and phishing URLs. We evaluate the effectiveness of the proposed method using a large set of one million URLs that incorporate both authentic and phishing URLs. Experimental results show that our model is effective, with an impressive accuracy of 97.75%, outperforming the baseline models. This study significantly improves online security by offering a novel and highly accurate phishing URL detection method.
{"title":"Generative adversarial network-based phishing URL detection with variational autoencoder and transformer","authors":"Jishnu Kaitholikkal Sasi, Arthi Balakrishnan","doi":"10.11591/ijai.v13.i2.pp2165-2172","DOIUrl":"https://doi.org/10.11591/ijai.v13.i2.pp2165-2172","url":null,"abstract":"Phishing attacks pose a constant threat to online security, necessitating the development of efficient tools for identifying malicious URLs. In this article, we propose a novel approach to detect phishing URLs employing a generative adversarial network (GAN) with a variational autoencoder (VAE) as the generator and a transformer model with self-attention as the discriminator. The VAE generator is trained to produce synthetic URLs. In contrast, the transformer discriminator uses its self-attention mechanism to focus on the different parts of the input URLs to extract crucial features. Our model uses adversarial training to distinguish between legitimate and phishing URLs. We evaluate the effectiveness of the proposed method using a large set of one million URLs that incorporate both authentic and phishing URLs. Experimental results show that our model is effective, with an impressive accuracy of 97.75%, outperforming the baseline models. This study significantly improves online security by offering a novel and highly accurate phishing URL detection method.","PeriodicalId":507934,"journal":{"name":"IAES International Journal of Artificial Intelligence (IJ-AI)","volume":"59 15","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141231612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：