Pub Date : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp794-805
Heba Mohammed Fadhil, Mohammed Abdullah, Mohammed Younis
In combinatorial testing development, the fabrication of covering arrays is the key challenge by the multiple aspects that influence it. A wide range of combinatorial problems can be solved using metaheuristic and greedy techniques. Combining the greedy technique utilizing a metaheuristic search technique like hill climbing (HC), can produce feasible results for combinatorial tests. Methods based on metaheuristics are used to deal with tuples that may be left after redundancy using greedy strategies; then the result utilization is assured to be near-optimal using a metaheuristic algorithm. As a result, the use of both greedy and HC algorithms in a single test generation system is a good candidate if constructed correctly. This study presents a hybrid greedy hill climbing algorithm (HGHC) that ensures both effectiveness and near-optimal results for generating a small number of test data. To make certain that the suggested HGHC outperforms the most used techniques in terms of test size. It is compared to others in order to determine its effectiveness. In contrast to recent practices utilized for the production of covering arrays (CAs) and mixed covering arrays (MCAs), this hybrid strategy is superior since allowing it to provide the utmost outcome while reducing the size and limit the loss of unique pairings in the CA/MCA generation.
{"title":"Innovations in t-way test creation based on a hybrid hill climbing-greedy algorithm","authors":"Heba Mohammed Fadhil, Mohammed Abdullah, Mohammed Younis","doi":"10.11591/ijai.v12.i2.pp794-805","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp794-805","url":null,"abstract":"<p>In combinatorial testing development, the fabrication of covering arrays is the key challenge by the multiple aspects that influence it. A wide range of combinatorial problems can be solved using metaheuristic and greedy techniques. Combining the greedy technique utilizing a metaheuristic search technique like hill climbing (HC), can produce feasible results for combinatorial tests. Methods based on metaheuristics are used to deal with tuples that may be left after redundancy using greedy strategies; then the result utilization is assured to be near-optimal using a metaheuristic algorithm. As a result, the use of both greedy and HC algorithms in a single test generation system is a good candidate if constructed correctly. This study presents a hybrid greedy hill climbing algorithm (HGHC) that ensures both effectiveness and near-optimal results for generating a small number of test data. To make certain that the suggested HGHC outperforms the most used techniques in terms of test size. It is compared to others in order to determine its effectiveness. In contrast to recent practices utilized for the production of covering arrays (CAs) and mixed covering arrays (MCAs), this hybrid strategy is superior since allowing it to provide the utmost outcome while reducing the size and limit the loss of unique pairings in the CA/MCA generation.</p>","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136261488","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp815-822
Akara Thammastitkul, Jitsanga Petsuwan
Thai Jasmine rice or Thai Hom Mali rice is a well-known rice type that originated in Thailand. Rice grain qualities are important in determining market pricing and are used in grading systems. The purpose of this research is to use machine learning and deep learning to improve the grading of Thai Hom Mali rice following standardized grading criteria. The appearance of grains and foreign items will determine the grade of rice. The experiment has two parts: grain categorization and rice grading. Multi-class support vector machine (SVM) and convolutional neural network (CNN) are proposed. There are 15 features used as input for multi-class SVM, including morphology and color features. With ImageNet pre-trained weights, CNN with DenseNet201 architecture is implemented. The experiment also tested into how CNN worked with both original and preprocessed images. The results are then compared to a neural network (NN) baseline approach. The CNN approach, which identified each rice variety using preprocessed images, archieved the greatest accuracy rate of 98.25%, with an average accuracy of 94.52% across six categories of rice grading.
{"title":"Thai Hom Mali rice grading using machine learning and deep learning approaches","authors":"Akara Thammastitkul, Jitsanga Petsuwan","doi":"10.11591/ijai.v12.i2.pp815-822","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp815-822","url":null,"abstract":"Thai Jasmine rice or Thai Hom Mali rice is a well-known rice type that originated in Thailand. Rice grain qualities are important in determining market pricing and are used in grading systems. The purpose of this research is to use machine learning and deep learning to improve the grading of Thai Hom Mali rice following standardized grading criteria. The appearance of grains and foreign items will determine the grade of rice. The experiment has two parts: grain categorization and rice grading. Multi-class support vector machine (SVM) and convolutional neural network (CNN) are proposed. There are 15 features used as input for multi-class SVM, including morphology and color features. With ImageNet pre-trained weights, CNN with DenseNet201 architecture is implemented. The experiment also tested into how CNN worked with both original and preprocessed images. The results are then compared to a neural network (NN) baseline approach. The CNN approach, which identified each rice variety using preprocessed images, archieved the greatest accuracy rate of 98.25%, with an average accuracy of 94.52% across six categories of rice grading.","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44785140","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp610-617
H. Pardede, Purwoko Adhi, Vicky Zilvan, A. Ramdan, Dikdik Krisnandi
There are great interests in developing speech recognition using deep learning technologies due to their capability to model the complexity of pronunciations, syntax, and language rules of speech data better than the traditional hidden Markov model (HMM) do. But, the availability of large amount of data is necessary for deep learning-based speech recognition to be effective. While this is not a problem for mainstream languages such as English or Chinese, this is not the case for non-mainstream languages such as Indonesian. To overcome this limitation, we present deep features based on convolutional neural networks (CNN) for Indonesian large vocabulary continuous speech recognition in this paper. The CNN is trained discriminatively which is different from usual deep learning implementations where the networks are trained generatively. Our evaluations show that the proposed method on Indonesian speech data achieves 7.26% and 9.01% error reduction rates over the state-of-the-art deep belief networks-deep neural networks (DBN-DNN) for large vocabulary continuous speech recognition (LVCSR), with Mel frequency cepstral coefficients (MFCC) and filterbank (FBANK) used as features, respectively. An error reduction rate of 6.13% is achieved compared to CNN-DNN with generative training.
{"title":"Deep convolutional neural networks-based features for Indonesian large vocabulary speech recognition","authors":"H. Pardede, Purwoko Adhi, Vicky Zilvan, A. Ramdan, Dikdik Krisnandi","doi":"10.11591/ijai.v12.i2.pp610-617","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp610-617","url":null,"abstract":"There are great interests in developing speech recognition using deep learning technologies due to their capability to model the complexity of pronunciations, syntax, and language rules of speech data better than the traditional hidden Markov model (HMM) do. But, the availability of large amount of data is necessary for deep learning-based speech recognition to be effective. While this is not a problem for mainstream languages such as English or Chinese, this is not the case for non-mainstream languages such as Indonesian. To overcome this limitation, we present deep features based on convolutional neural networks (CNN) for Indonesian large vocabulary continuous speech recognition in this paper. The CNN is trained discriminatively which is different from usual deep learning implementations where the networks are trained generatively. Our evaluations show that the proposed method on Indonesian speech data achieves 7.26% and 9.01% error reduction rates over the state-of-the-art deep belief networks-deep neural networks (DBN-DNN) for large vocabulary continuous speech recognition (LVCSR), with Mel frequency cepstral coefficients (MFCC) and filterbank (FBANK) used as features, respectively. An error reduction rate of 6.13% is achieved compared to CNN-DNN with generative training.","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"65349293","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp585-592
R. Jiménez-Moreno, Ricardo A. Castillo
This work presents the design and validation of a voice assistant to command robotic tasks in a residential environment, as a support for people who require isolation or support due to body motor problems. The preprocessing of a database of 3600 audios of 8 different categories of words like “paper”, “glass” or “robot”, that allow to conform commands such as "carry paper" or "bring medicine", obtaining a matrix array of Mel frequencies and its derivatives, as inputs to a convolutional neural network that presents an accuracy of 96.9% in the discrimination of the categories. The command recognition tests involve recognizing groups of three words starting with "robot", for example, "robot bring glass", and allow identifying 8 different actions per voice command, with an accuracy of 88.75%.
{"title":"Deep learning speech recognition for residential assistant robot","authors":"R. Jiménez-Moreno, Ricardo A. Castillo","doi":"10.11591/ijai.v12.i2.pp585-592","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp585-592","url":null,"abstract":"This work presents the design and validation of a voice assistant to command robotic tasks in a residential environment, as a support for people who require isolation or support due to body motor problems. The preprocessing of a database of 3600 audios of 8 different categories of words like “paper”, “glass” or “robot”, that allow to conform commands such as \"carry paper\" or \"bring medicine\", obtaining a matrix array of Mel frequencies and its derivatives, as inputs to a convolutional neural network that presents an accuracy of 96.9% in the discrimination of the categories. The command recognition tests involve recognizing groups of three words starting with \"robot\", for example, \"robot bring glass\", and allow identifying 8 different actions per voice command, with an accuracy of 88.75%.","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47806239","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp831-839
Malak Abdullah, M. Al-Ayyoub, Farah Shatnawi, Saif Rawashdeh, Rob Abbott
The outbreak of coronavirus disease 2019 (COVID-19) drives most higher education systems in many countries to stop face-to-face learning. Accordingly, many universities, including Jordan University of Science and Technology (JUST), changed the teaching method from face-to-face education to electronic learning from a distance. This research paper investigated the impact of the e-learning experience on the students during the spring semester of 2020 at JUST. It also explored how to predict students’ academic performances using e-learning data. Consequently, we collected students’ datasets from two resources: the center for e-learning and open educational resources and the admission and registration unit at the university. Five courses in the spring semester of 2020 were targeted. In addition, four regression machine learning algorithms had been used in this study to generate the predictions: random forest (RF), Bayesian ridge (BR), adaptive boosting (AdaBoost), and extreme gradient boosting (XGBoost). The results showed that the ensemble model for RF and XGBoost yielded the best performance. Finally, it is worth mentioning that among all the e-learning components and events, quiz events had a significant impact on predicting the student’s academic performance. Moreover, the paper shows that the activities between weeks 9 and 12 influenced students’ performances during the semester.
{"title":"Predicting students’ academic performance using e-learning logs","authors":"Malak Abdullah, M. Al-Ayyoub, Farah Shatnawi, Saif Rawashdeh, Rob Abbott","doi":"10.11591/ijai.v12.i2.pp831-839","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp831-839","url":null,"abstract":"The outbreak of coronavirus disease 2019 (COVID-19) drives most higher education systems in many countries to stop face-to-face learning. Accordingly, many universities, including Jordan University of Science and Technology (JUST), changed the teaching method from face-to-face education to electronic learning from a distance. This research paper investigated the impact of the e-learning experience on the students during the spring semester of 2020 at JUST. It also explored how to predict students’ academic performances using e-learning data. Consequently, we collected students’ datasets from two resources: the center for e-learning and open educational resources and the admission and registration unit at the university. Five courses in the spring semester of 2020 were targeted. In addition, four regression machine learning algorithms had been used in this study to generate the predictions: random forest (RF), Bayesian ridge (BR), adaptive boosting (AdaBoost), and extreme gradient boosting (XGBoost). The results showed that the ensemble model for RF and XGBoost yielded the best performance. Finally, it is worth mentioning that among all the e-learning components and events, quiz events had a significant impact on predicting the student’s academic performance. Moreover, the paper shows that the activities between weeks 9 and 12 influenced students’ performances during the semester.","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" 40","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41253178","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp861-873
S. S. Samaan, H. A. Jeiad
Precise traffic classification is essential to numerous network functionalities such as routing, network management, and resource allocation. Traditional classification techniques became insufficient due to the massive growth of network traffic that requires high computational costs. The arising model of software defined networking (SDN) has adjusted the network architecture to get a centralized controller that preserves a global view over the entire network. This paper proposes a model for SDN traffic classification based on machine learning (ML) using the Spark framework. The proposed model consists of two phases; learning and deployment. A ML pipeline is constructed in the learning phase, consisting of a set of stages combined as a single entity. Three ML models are built and evaluated; decision tree, random forest, and logistic regression, for classifying a well-known 75 applications, including Google and YouTube, accurately and in a short time scale. A dataset consisting of 3,577,296 flows with 87 features is used for training and testing the models. The decision tree model is elected for deployment according to the performance results, which indicate that it has the best accuracy with 0.98. The performance of the proposed model is compared with the state-of-the-art works, and better accuracy result is reported.
{"title":"Architecting a machine learning pipeline for online traffic classification in software defined networking using spark","authors":"S. S. Samaan, H. A. Jeiad","doi":"10.11591/ijai.v12.i2.pp861-873","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp861-873","url":null,"abstract":"Precise traffic classification is essential to numerous network functionalities such as routing, network management, and resource allocation. Traditional classification techniques became insufficient due to the massive growth of network traffic that requires high computational costs. The arising model of software defined networking (SDN) has adjusted the network architecture to get a centralized controller that preserves a global view over the entire network. This paper proposes a model for SDN traffic classification based on machine learning (ML) using the Spark framework. The proposed model consists of two phases; learning and deployment. A ML pipeline is constructed in the learning phase, consisting of a set of stages combined as a single entity. Three ML models are built and evaluated; decision tree, random forest, and logistic regression, for classifying a well-known 75 applications, including Google and YouTube, accurately and in a short time scale. A dataset consisting of 3,577,296 flows with 87 features is used for training and testing the models. The decision tree model is elected for deployment according to the performance results, which indicate that it has the best accuracy with 0.98. The performance of the proposed model is compared with the state-of-the-art works, and better accuracy result is reported.","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44308274","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 : 2023-06-01DOI: 10.11591/ijai.v12.i2.pp747-754
Sathishkumar Mani, Reshmy Avanavalappil Krishnankutty, Sabaria Swaminathan, P. Theerthagiri
Quality is the most determining factor for any product. Optimal care and best measures are to be taken in assessing the quality of any product. This work deals with determining the quality of wine using intelligence-based learning techniques. In order to estimate the quality of wine, several experiments are performed on wine datasets. The main purpose of our work is to study and discover an efficient machine learning (ML) model that could determine the quality of wine given some Physico-chemical features. This study establishes that selecting important features to evaluate rather than all of them can lead to improved forecasts. According to the results, this approach may provide people who are not wine experts a greater opportunity to choose a fine wine.
{"title":"An investigation of wine quality testing using machine learning techniques","authors":"Sathishkumar Mani, Reshmy Avanavalappil Krishnankutty, Sabaria Swaminathan, P. Theerthagiri","doi":"10.11591/ijai.v12.i2.pp747-754","DOIUrl":"https://doi.org/10.11591/ijai.v12.i2.pp747-754","url":null,"abstract":"<div class=\"page\" title=\"Page 1\"><div class=\"layoutArea\"><div class=\"column\"><p><span>Quality is the most determining factor for any product. Optimal care and best measures are to be taken in assessing the quality of any product. This work deals with determining the quality of wine using intelligence-based learning techniques. In order to estimate the quality of wine, several experiments are performed on wine datasets. The main purpose of our work is to study and discover an efficient machine learning (ML) model that could determine the quality of wine given some Physico-chemical features. This study establishes that selecting important features to evaluate rather than all of them can lead to improved forecasts. According to the results, this approach may provide people who are not wine experts a greater opportunity to choose a fine wine.</span></p></div></div></div>","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45458507","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 : 2023-03-01DOI: 10.11591/ijai.v12.i1.pp384-393
Cheng Xiao Ge, M. A. As’ari, Nur Anis Jasmin Sufri
The coronavirus disease 2019 (COVID-19) is a highly infectious disease caused by the SARS-CoV-2 coronavirus. In breaking the transmission chain of SARS-CoV-2, the government has made it compulsory for the people to wear a mask in public places to prevent COVID-19 transmission. Hence, an automated face mask detection is crucial to facilitate the monitoring process in ensuring people to wear a face mask in public. This project aims to develop an automated face and face mask detection for multiple people by applying deep learning-based object detection algorithm you only look once version 3 (YOLOv3). YOLOv3 object detection algorithm was concatenated with different backbones including ResNet-50 and Darknet-53 to develop the face and face mask detection model. Datasets were collected from online resources including Kaggle and Github and the images were filtered and labelled accordingly. The models were trained on 4393 images and evaluated based on precision, recall, mean average precision and the detection time. In conclusion, DarkNet53_YOLOv3 was chosen as the better model compared to ResNet50_YOLOv3 model with its good performance on accuracy with a mAP of 95.94% and a fast detection speed with a detection time of 50 seconds on 776 images.
{"title":"Multiple face mask wearer detection based on YOLOv3 approach","authors":"Cheng Xiao Ge, M. A. As’ari, Nur Anis Jasmin Sufri","doi":"10.11591/ijai.v12.i1.pp384-393","DOIUrl":"https://doi.org/10.11591/ijai.v12.i1.pp384-393","url":null,"abstract":"The coronavirus disease 2019 (COVID-19) is a highly infectious disease caused by the SARS-CoV-2 coronavirus. In breaking the transmission chain of SARS-CoV-2, the government has made it compulsory for the people to wear a mask in public places to prevent COVID-19 transmission. Hence, an automated face mask detection is crucial to facilitate the monitoring process in ensuring people to wear a face mask in public. This project aims to develop an automated face and face mask detection for multiple people by applying deep learning-based object detection algorithm you only look once version 3 (YOLOv3). YOLOv3 object detection algorithm was concatenated with different backbones including ResNet-50 and Darknet-53 to develop the face and face mask detection model. Datasets were collected from online resources including Kaggle and Github and the images were filtered and labelled accordingly. The models were trained on 4393 images and evaluated based on precision, recall, mean average precision and the detection time. In conclusion, DarkNet53_YOLOv3 was chosen as the better model compared to ResNet50_YOLOv3 model with its good performance on accuracy with a mAP of 95.94% and a fast detection speed with a detection time of 50 seconds on 776 images. ","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49365759","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 : 2023-03-01DOI: 10.11591/ijai.v12.i1.pp79-86
Komala Karilingappa, D. Jayadevappa, Shivaprakash Ganganna
Facial expression is a kind of nonverbal communication that conveys information about a person's emotional state. Human emotion detection and recognition remains a major task in computer vision (CV) and artificial intelligence (AI). To recognize and identify the many sorts of emotions, several algorithms are proposed in the literature. In this paper, the modified Viola-Jones method is introduced to provide a robust approach capable of detecting and identifying human feelings such as angerness,sadness, desire, surprise, anxiety, disgust, and neutrality in real-time. This technique captures real-time pictures and then extracts the characteristics of the facial image to identify emotions very accurately. In this method, many feature extraction techniques like gray-level co-occurrence matrix (GLCM), linear binary pattern (LBP) and robust principal components analysis (RPCA) are applied to identify the distinct mood states and they are categorized using a convolution neural network (CNN) classifier. The obtained outcome demonstrates that the proposed method outperforms in terms of determining the rate of emotion recognition as compared to the current human emotion recognition techniques.
{"title":"Human emotion detection and classification using modified viola-jones and convolution neural network","authors":"Komala Karilingappa, D. Jayadevappa, Shivaprakash Ganganna","doi":"10.11591/ijai.v12.i1.pp79-86","DOIUrl":"https://doi.org/10.11591/ijai.v12.i1.pp79-86","url":null,"abstract":"<span lang=\"EN-US\">Facial expression is a kind of nonverbal communication that conveys information about a person's emotional state. Human emotion detection and recognition remains a major task in computer vision (CV) and artificial intelligence (AI). To recognize and identify the many sorts of emotions, several algorithms are proposed in the literature. In this paper, the modified Viola-Jones method is introduced to provide a robust approach capable of detecting and identifying human feelings such as angerness,sadness, desire, surprise, anxiety, disgust, and neutrality in real-time. This technique captures real-time pictures and then extracts the characteristics of the facial image to identify emotions very accurately. In this method, many feature extraction techniques like gray-level co-occurrence matrix (GLCM), linear binary pattern (LBP) and robust principal components analysis (RPCA) are applied to identify the distinct mood states and they are categorized using a convolution neural network (CNN) classifier. The obtained outcome demonstrates that the proposed method outperforms in terms of determining the rate of emotion recognition as compared to the current human emotion recognition techniques.</span><br /><div style=\"mso-element: comment-list;\"><div style=\"mso-element: comment;\"><div id=\"_com_1\" class=\"msocomtxt\"><!--[if !supportAnnotations]--></div><!--[endif]--></div></div>","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46284653","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 : 2023-03-01DOI: 10.11591/ijai.v12.i1.pp284-294
Mohammed Maree, Mujahed Eleyat, Shatha Rabayah, M. Belkhatir
Current lexica and machine learning based sentiment analysis approaches still suffer from a two-fold limitation. First, manual lexicon construction and machine training is time consuming and error-prone. Second, the prediction’s accuracy entails sentences and their corresponding training text should fall under the same domain. In this article, we experimentally evaluate four sentiment classifiers, namely Support Vector Machines, Naive Bayes, Logistic Regression and Random Forest. We quantify the quality of each of these models using three real-world datasets that comprise 50,000 movie reviews, 10,662 sentences, and 300 generic movie reviews. Specifically, we study the impact of a variety of natural language processing (NLP) pipelines on the quality of the predicted sentiment orientations. Additionally, we measure the impact of incorporating lexical semantic knowledge captured by WordNet on expanding original words in sentences. Findings demonstrate that the utilizing different NLP pipelines and semantic relationships impacts the quality of the sentiment analyzers. In particular, results indicate that coupling lemmatization and knowledge-based n-gram features proved to produce higher accuracy results. With this coupling, the accuracy of the support vector machine (SVM) classifier has improved to 90.43%, while it was 86.83%, 90.11%, 86.20%, respectively using the three other classifiers.
{"title":"A hybrid composite features based sentence level sentiment analyzer","authors":"Mohammed Maree, Mujahed Eleyat, Shatha Rabayah, M. Belkhatir","doi":"10.11591/ijai.v12.i1.pp284-294","DOIUrl":"https://doi.org/10.11591/ijai.v12.i1.pp284-294","url":null,"abstract":"Current lexica and machine learning based sentiment analysis approaches still suffer from a two-fold limitation. First, manual lexicon construction and machine training is time consuming and error-prone. Second, the prediction’s accuracy entails sentences and their corresponding training text should fall under the same domain. In this article, we experimentally evaluate four sentiment classifiers, namely Support Vector Machines, Naive Bayes, Logistic Regression and Random Forest. We quantify the quality of each of these models using three real-world datasets that comprise 50,000 movie reviews, 10,662 sentences, and 300 generic movie reviews. Specifically, we study the impact of a variety of natural language processing (NLP) pipelines on the quality of the predicted sentiment orientations. Additionally, we measure the impact of incorporating lexical semantic knowledge captured by WordNet on expanding original words in sentences. Findings demonstrate that the utilizing different NLP pipelines and semantic relationships impacts the quality of the sentiment analyzers. In particular, results indicate that coupling lemmatization and knowledge-based n-gram features proved to produce higher accuracy results. With this coupling, the accuracy of the support vector machine (SVM) classifier has improved to 90.43%, while it was 86.83%, 90.11%, 86.20%, respectively using the three other classifiers. ","PeriodicalId":52221,"journal":{"name":"IAES International Journal of Artificial Intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46350228","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号