Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32804
Xing Ma
Seaborne crude oil remains the main source of energy in the modern world in terms of volume, accounting for nearly half of all internationally traded crude oil. The shipping market is already characterized by high volatility, coupled with the impact of COVID-19 lockdown and geopolitics events. Price forecasting has become a necessary and challenging task for shipowners and other stakeholders. In the shipping market forecasting literature, the usual focus is on the newbuilding ship price or freight rate. A limited number of literature is for secondhand tanker price. On the other hand, there is few literature that use wavelet neural networks based on adaptive genetic algorithm (AGA-WNN) to predict shipping market. This paper mainly studies the application of the hybrid model to secondhand price prediction of 5 kinds of tanker sizes. The performance of AGA-WNN on time series of 10 and 15 years is compared with the basic performance provided by the six benchmark models, using three error metrics and two statistical tests. We can point out that AGA-WNN provides encouraging and promising results, outperforming the baseline models in both accuracy and robustness. It can be said that AGA-WNN gives the best overall predictive performance.
{"title":"Forecasting Secondhand Tanker Price Through Wavelet Neural Networks Based on Adaptive Genetic Algorithm","authors":"Xing Ma","doi":"10.5755/j01.itc.52.2.32804","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32804","url":null,"abstract":"Seaborne crude oil remains the main source of energy in the modern world in terms of volume, accounting for nearly half of all internationally traded crude oil. The shipping market is already characterized by high volatility, coupled with the impact of COVID-19 lockdown and geopolitics events. Price forecasting has become a necessary and challenging task for shipowners and other stakeholders. In the shipping market forecasting literature, the usual focus is on the newbuilding ship price or freight rate. A limited number of literature is for secondhand tanker price. On the other hand, there is few literature that use wavelet neural networks based on adaptive genetic algorithm (AGA-WNN) to predict shipping market. This paper mainly studies the application of the hybrid model to secondhand price prediction of 5 kinds of tanker sizes. The performance of AGA-WNN on time series of 10 and 15 years is compared with the basic performance provided by the six benchmark models, using three error metrics and two statistical tests. We can point out that AGA-WNN provides encouraging and promising results, outperforming the baseline models in both accuracy and robustness. It can be said that AGA-WNN gives the best overall predictive performance.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"25 1","pages":"336-357"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83255434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.33214
Zhi Li, Chaofeng Li, Tuxin Guan, Shaopeng Shang
Underwater object detection is one of the important technologies for improving the efficiency of underwater inspection, but the existing methods still suffer from the problems of missed detection and insufficient target localization capability of targets. To address these problems, an improved Transformer and multi-scale attentional supervised feature fusion-based underwater object detection method is proposed. In our method, the underwater objects are preprocessed by prior knowledge first. Then, a new coordinate decomposition window-based (CDW) Transformer block is proposed to extract spatial location information more accurately, and scaling factors are introduced to reduce the intermediate computation. Finally, an attentional supervised fusion (ASF) method is proposed to strengthen the link between feature extraction and feature fusion, and further improve the detected performance by using compound attention weights. The cascade detection head is improved, where the information flow is reversed to enhance the prediction of coordinates. The average accuracy of the proposed method on the URPC and DUO datasets is 3.7% and 3.8% higher than that of the baseline network through the cross-test, and outperforms the state-of-the-art methods. This study can provide a reference for engineering applications such as automated marine operations and biodetected fishing techniques.
{"title":"Underwater Object Detection Based on Improved Transformer and Attentional Supervised Fusion","authors":"Zhi Li, Chaofeng Li, Tuxin Guan, Shaopeng Shang","doi":"10.5755/j01.itc.52.2.33214","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33214","url":null,"abstract":"Underwater object detection is one of the important technologies for improving the efficiency of underwater inspection, but the existing methods still suffer from the problems of missed detection and insufficient target localization capability of targets. To address these problems, an improved Transformer and multi-scale attentional supervised feature fusion-based underwater object detection method is proposed. In our method, the underwater objects are preprocessed by prior knowledge first. Then, a new coordinate decomposition window-based (CDW) Transformer block is proposed to extract spatial location information more accurately, and scaling factors are introduced to reduce the intermediate computation. Finally, an attentional supervised fusion (ASF) method is proposed to strengthen the link between feature extraction and feature fusion, and further improve the detected performance by using compound attention weights. The cascade detection head is improved, where the information flow is reversed to enhance the prediction of coordinates. The average accuracy of the proposed method on the URPC and DUO datasets is 3.7% and 3.8% higher than that of the baseline network through the cross-test, and outperforms the state-of-the-art methods. This study can provide a reference for engineering applications such as automated marine operations and biodetected fishing techniques.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"34 1","pages":"397-415"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76833367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aspect-category-opinion-sentiment (ACOS) quadruples play an essential role in implicit sentiment analysis. Considering the distances between aspects and opinions in sentences, a novel Distance-Extract-Classify-ACOS quadruple extraction method with distance information between aspects and opinions is proposed. Compared with Double-Propagation-ACOS, JET-BERT-ACOS and Extract-Classify-ACOS quadruple extraction models, the recall and F1 scores of the Distance-Extract-Classify-ACOS quadruple extraction model respectively increase by 2.08%-35.81% and 1.47%-36.7% on the Restaurant-ACOS and Laptop-ACOS datasets. Using the extracted quadruples for implicit sentiment analysis, the performance of the LSTM, GRU, TextCNN, and BERT models significantly outperforms these models with original sentences, aspects-opinions pairs, and aspects-categories-opinions triples on Restaurant-ACOS and Laptop-ACOS datasets.
{"title":"An Aspect-Category-Opinion-Sentiment Quadruple Extraction with Distance Information for Implicit Sentiment Analysis","authors":"Jianwei Li, Xianyong Li, Yajun Du, Yongquan Fan, Xiaoliang Chen, Dong Huang","doi":"10.5755/j01.itc.52.2.32903","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32903","url":null,"abstract":"The aspect-category-opinion-sentiment (ACOS) quadruples play an essential role in implicit sentiment analysis. Considering the distances between aspects and opinions in sentences, a novel Distance-Extract-Classify-ACOS quadruple extraction method with distance information between aspects and opinions is proposed. Compared with Double-Propagation-ACOS, JET-BERT-ACOS and Extract-Classify-ACOS quadruple extraction models, the recall and F1 scores of the Distance-Extract-Classify-ACOS quadruple extraction model respectively increase by 2.08%-35.81% and 1.47%-36.7% on the Restaurant-ACOS and Laptop-ACOS datasets. Using the extracted quadruples for implicit sentiment analysis, the performance of the LSTM, GRU, TextCNN, and BERT models significantly outperforms these models with original sentences, aspects-opinions pairs, and aspects-categories-opinions triples on Restaurant-ACOS and Laptop-ACOS datasets.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"8 1","pages":"445-456"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75946338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32803
Guanyu Ren
Monkeypox has been recognized as the next global pandemic after COVID-19 and its potential damage cannot be neglected. Computer vision-based diagnosis and detection method with deep learning models have been proven effective during the COVID-19 period. However, with limited samples, the deep learning models are difficult to be full trained. In this paper, twelve CNN-based models, including VGG16, VGG19, ResNet152, DenseNet121, DenseNet201, EfficientNetB7, EfficientNetV2B3, EfficientNetV2M and InceptionV3, are used for monkeypox detection with limited skin pictures. Numerical results suggest that DenseNet201 achieves the best classification accuracy of 98.89% for binary classification, 100% for four-class classification and 99.94% for six-class classification over the rest models.
{"title":"Monkeypox Disease Detection with Pretrained Deep Learning Models","authors":"Guanyu Ren","doi":"10.5755/j01.itc.52.2.32803","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32803","url":null,"abstract":"Monkeypox has been recognized as the next global pandemic after COVID-19 and its potential damage cannot be neglected. Computer vision-based diagnosis and detection method with deep learning models have been proven effective during the COVID-19 period. However, with limited samples, the deep learning models are difficult to be full trained. In this paper, twelve CNN-based models, including VGG16, VGG19, ResNet152, DenseNet121, DenseNet201, EfficientNetB7, EfficientNetV2B3, EfficientNetV2M and InceptionV3, are used for monkeypox detection with limited skin pictures. Numerical results suggest that DenseNet201 achieves the best classification accuracy of 98.89% for binary classification, 100% for four-class classification and 99.94% for six-class classification over the rest models.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"11 1","pages":"288-296"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84359912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.33302
Satheesh Pandian, A. Kalpana
To monitor electrical indications from the heart and assess its performance, the electrocardiogram (ECG) is the most common and routine diagnostic instrument employed. Cardiac arrhythmias are only one example of the many heart conditions people might have. ECG records are used to diagnose an arrhythmia, an abnormal cardiac beat that can cause a stroke in extreme circumstances. However, due to the extensive data that an ECG contains, it is quite difficult to glean the necessary information through visual analysis. Therefore, it is crucial to develop an effective (automatic) method to analyze the vast amounts of data available from ECG. For decades, researchers have focused on developing methods to automatically and computationally categorize and identify cardiac arrhythmias. However, monitoring for arrhythmias in real-time is challenging. To streamline the detection and classification process, this research presents a hybrid deep learning-based technique. There are two major contributions to this study. To automate the noise reduction and feature extraction, 1D ECG data are first transformed into 2D Scalogram images. Following this, a combined approach called the Residual attention-based 2D-CNN-LSTM-CNN (RACLC) is recommended by merging multiple learning models, specifically the 2D convolutional neural network (CNN) and the Long Short-Term Memory (LSTM) system, based on research findings. The name of this model comes from a combination of the two deep learning. Both the beats themselves, which provide morphological information, and the beats paired with neighboring segments, which provide temporal information, are essential. Our suggested model simultaneously collects time-domain and morphological ECG signal data and combines them. The application of the attention block to the network helps to strengthen the valuable information, acquire the confidential message in the ECG signal, and boost the efficiency of the model when it comes to categorization. To evaluate the efficacy of the proposed RACLC method, we carried out a complete experimental investigation making use of the MIT-BIH arrhythmia database, which is used by a large number of researchers. The results of our experiments show that the automated detection method we propose is effective.
{"title":"HybDeepNet: ECG Signal Based Cardiac Arrhythmia Diagnosis Using a Hybrid Deep Learning Model","authors":"Satheesh Pandian, A. Kalpana","doi":"10.5755/j01.itc.52.2.33302","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33302","url":null,"abstract":"To monitor electrical indications from the heart and assess its performance, the electrocardiogram (ECG) is the most common and routine diagnostic instrument employed. Cardiac arrhythmias are only one example of the many heart conditions people might have. ECG records are used to diagnose an arrhythmia, an abnormal cardiac beat that can cause a stroke in extreme circumstances. However, due to the extensive data that an ECG contains, it is quite difficult to glean the necessary information through visual analysis. Therefore, it is crucial to develop an effective (automatic) method to analyze the vast amounts of data available from ECG. For decades, researchers have focused on developing methods to automatically and computationally categorize and identify cardiac arrhythmias. However, monitoring for arrhythmias in real-time is challenging. To streamline the detection and classification process, this research presents a hybrid deep learning-based technique. There are two major contributions to this study. To automate the noise reduction and feature extraction, 1D ECG data are first transformed into 2D Scalogram images. Following this, a combined approach called the Residual attention-based 2D-CNN-LSTM-CNN (RACLC) is recommended by merging multiple learning models, specifically the 2D convolutional neural network (CNN) and the Long Short-Term Memory (LSTM) system, based on research findings. The name of this model comes from a combination of the two deep learning. Both the beats themselves, which provide morphological information, and the beats paired with neighboring segments, which provide temporal information, are essential. Our suggested model simultaneously collects time-domain and morphological ECG signal data and combines them. The application of the attention block to the network helps to strengthen the valuable information, acquire the confidential message in the ECG signal, and boost the efficiency of the model when it comes to categorization. To evaluate the efficacy of the proposed RACLC method, we carried out a complete experimental investigation making use of the MIT-BIH arrhythmia database, which is used by a large number of researchers. The results of our experiments show that the automated detection method we propose is effective.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"66 1","pages":"416-432"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80000766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32899
S. Karthikeyini, G. Vidhya, T. Vetriselvi, K. Deepa
In recent years, heart disease has superseded several other contributory death factors. It is challenging to predict an individual’s risk of acquiring heart disease since it requires both expert knowledge and real-world experience. Developing an effective method for the prognosis of heart disease using Internet of Medical Things (IoMT) technology in healthcare organizations by collecting sensor data from patients’ bodies, utilizing robust expert systems, and incorporating vast healthcare data on cardiac disorders to alert physicians in critical situations is a challenging task. Several machine learning-based techniques for predicting and diagnosing cardiac disease have recently been demonstrated. However, these algorithms cannot effectively handle high-dimensionalinformation due to the need for an intelligent framework incorporating multiple sources to predict cardiac illness. This work proposes a unique model for heart disease prediction based on deep learning, Deep Gated Recurrent Units (D-GRU), which combines with Stacked Auto Encoders. A novel optimization algorithm, the Logistic Chaos Honey Badger Algorithm, is proposed for optimal feature selection. Publicly available heart disease-related datasets collected from UCI Repository, Cleveland Database, are used for training the proposed D-GRU model. The trained model is further tested on the data gathered from the sensors in the IoMT framework. The performance of the proposed model is compared against several deep learning models and existing works in the literature. The proposed D-GRU model outperforms the other models taken for comparison andexhibits performance supremacy with an accuracy of 95.15% in predicting heart diseases.
{"title":"Heart Disease Prognosis Using D-GRU with Logistic Chaos Honey Badger Optimization in IoMT Framework","authors":"S. Karthikeyini, G. Vidhya, T. Vetriselvi, K. Deepa","doi":"10.5755/j01.itc.52.2.32899","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32899","url":null,"abstract":"In recent years, heart disease has superseded several other contributory death factors. It is challenging to predict an individual’s risk of acquiring heart disease since it requires both expert knowledge and real-world experience. Developing an effective method for the prognosis of heart disease using Internet of Medical Things (IoMT) technology in healthcare organizations by collecting sensor data from patients’ bodies, utilizing robust expert systems, and incorporating vast healthcare data on cardiac disorders to alert physicians in critical situations is a challenging task. Several machine learning-based techniques for predicting and diagnosing cardiac disease have recently been demonstrated. However, these algorithms cannot effectively handle high-dimensionalinformation due to the need for an intelligent framework incorporating multiple sources to predict cardiac illness. This work proposes a unique model for heart disease prediction based on deep learning, Deep Gated Recurrent Units (D-GRU), which combines with Stacked Auto Encoders. A novel optimization algorithm, the Logistic Chaos Honey Badger Algorithm, is proposed for optimal feature selection. Publicly available heart disease-related datasets collected from UCI Repository, Cleveland Database, are used for training the proposed D-GRU model. The trained model is further tested on the data gathered from the sensors in the IoMT framework. The performance of the proposed model is compared against several deep learning models and existing works in the literature. The proposed D-GRU model outperforms the other models taken for comparison andexhibits performance supremacy with an accuracy of 95.15% in predicting heart diseases.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"17 1","pages":"367-380"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73825268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32993
R. Ram, J. Akilandeswari, M. V. Kumar
The problem to be addressed is the high mortality rate of heart disease and the need for reliable and early detection techniques to prevent fatalities. Several clinical tests, including electrocardiogram (ECG) signals, heart sound signals, impedance cardiography (ICG), magnetic resonance imaging, and computer tomography can be used to determine whether an individual has heart disease. In this research, three deep learning models - Multilayer Perceptrons (MLPs), Deep Belief Networks (DBNs), and Restricted Boltzmann Machines (RBMs) - were used to detect heart disease by using the electrocardiogram (ECG) signal as the primary source. The publicly available datasets MIT-BIH and PTB-ECG were used to train and validate the proposed model. The results showed that the proposed hybrid model achieved the best performance compared to existing models, with an accuracy of 98.6%, 97.4%, and 96.2% on the MIT-BIH dataset, and 97.1%, 96.4%, and 95.3% on the PTB-ECG dataset, respectively. Furthermore, the model had excellent F1-score and AUC values, indicating the robustness of the proposed approach.
{"title":"HybDeepNet: A Hybrid Deep Learning Model for Detecting Cardiac Arrhythmia from ECG Signals","authors":"R. Ram, J. Akilandeswari, M. V. Kumar","doi":"10.5755/j01.itc.52.2.32993","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32993","url":null,"abstract":"The problem to be addressed is the high mortality rate of heart disease and the need for reliable and early detection techniques to prevent fatalities. Several clinical tests, including electrocardiogram (ECG) signals, heart sound signals, impedance cardiography (ICG), magnetic resonance imaging, and computer tomography can be used to determine whether an individual has heart disease. In this research, three deep learning models - Multilayer Perceptrons (MLPs), Deep Belief Networks (DBNs), and Restricted Boltzmann Machines (RBMs) - were used to detect heart disease by using the electrocardiogram (ECG) signal as the primary source. The publicly available datasets MIT-BIH and PTB-ECG were used to train and validate the proposed model. The results showed that the proposed hybrid model achieved the best performance compared to existing models, with an accuracy of 98.6%, 97.4%, and 96.2% on the MIT-BIH dataset, and 97.1%, 96.4%, and 95.3% on the PTB-ECG dataset, respectively. Furthermore, the model had excellent F1-score and AUC values, indicating the robustness of the proposed approach.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"37 1","pages":"433-444"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73877915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.33077
Mahmoud Hossein Zadeh, I. Çiçekli
Protest Event Analysis is important for government officials and social scientists. Here we present a new method for predicting protest events and identifying indicators of protests and violence by monitoring the content generated on Twitter. By identifying these indicators, protests and the possibility of violence can be predicted and controlled more accurately. Twitter user behaviors such as opinion share and event log share are used as indicators and this study presents a new method based on a Bayesian logistic regression algorithm for predicting protests and violence using Twitter user behaviors. According to the proposed method, users’ event log share behaviors which include the rate of tweets containing date and time information is the reliable indicator for identifying protests. Users’ opinion share behaviors which include hate-anger tweet rates is also best for identifying violence in protests. A dataset which consists of tweets that are generated on protests in the Black Lives Matter (BLM) movement after the death of George Floyd is used in the evaluation of the proposed method. According to information published on acleddata.com, protests and violence have been reported in various cities on specific dates. The dataset contains 1414 protest events and 3078 non-protest events from 460 cities in 37 U.S. states. Protest events in the BLM movement between May 28 and June 30 among which 285 were violent and 1129 were peaceful. Our proposed method is tested on this dataset and the occurrence of protests is predicted with 85% precision. It is also possible to predict violence in protests with 85% precision with our method on this dataset. This study provides a successful method to predict small and large-scale protests, different from the existing literature focusing on large-scale protests.
{"title":"Protest Event Analysis: A New Method Based on Twitter's User Behaviors","authors":"Mahmoud Hossein Zadeh, I. Çiçekli","doi":"10.5755/j01.itc.52.2.33077","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33077","url":null,"abstract":"Protest Event Analysis is important for government officials and social scientists. Here we present a new method for predicting protest events and identifying indicators of protests and violence by monitoring the content generated on Twitter. By identifying these indicators, protests and the possibility of violence can be predicted and controlled more accurately. Twitter user behaviors such as opinion share and event log share are used as indicators and this study presents a new method based on a Bayesian logistic regression algorithm for predicting protests and violence using Twitter user behaviors. According to the proposed method, users’ event log share behaviors which include the rate of tweets containing date and time information is the reliable indicator for identifying protests. Users’ opinion share behaviors which include hate-anger tweet rates is also best for identifying violence in protests.\u0000A dataset which consists of tweets that are generated on protests in the Black Lives Matter (BLM) movement after the death of George Floyd is used in the evaluation of the proposed method. According to information published on acleddata.com, protests and violence have been reported in various cities on specific dates. The dataset contains 1414 protest events and 3078 non-protest events from 460 cities in 37 U.S. states. Protest events in the BLM movement between May 28 and June 30 among which 285 were violent and 1129 were peaceful. Our proposed method is tested on this dataset and the occurrence of protests is predicted with 85% precision. It is also possible to predict violence in protests with 85% precision with our method on this dataset. This study provides a successful method to predict small and large-scale protests, different from the existing literature focusing on large-scale protests.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"49 1","pages":"457-470"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75676984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32873
Tianwei Shi, Ke Chen, Ling Ren, Wenhua Cui
This paper puts forward a brain computer interface (BCI) system to realize the hand and wrist control using the ABB Mechanical Arm. This BCI system gathers four kinds of motor imaginary (MI) tasks (hand grasp, hand spread, wrist flexion and wrist extension) electroencephalogram (EEG) signals from 30 electrodes. It utilizes two fifth-order Butterworth Band-Pass Filter (BPF) with different bandwidths and normalization method to achieve the raw MI tasks EEG signals preprocessing. The main challenge of feature extraction is to extract enough representative features from MI tasks to classify them. This proposed BCI system extracts eleven kinds of features in time domain and time-frequency domain and uses mutual information method to reduce the large dimension of the extracted features. In addition, the BCI system applies a single convolutional layer Convolutional neural networks (CNN) with 30 filters to implement the quaternary classification of MI tasks. Compared with early researches, the classification accuracy of this BCI system is increased by about 35%. The actual mechanical arm grasping control experiments verifies that this BCI system has good adaptability.
{"title":"Brain Computer Interface Based on Motor Imagery for Mechanical Arm Grasp Control","authors":"Tianwei Shi, Ke Chen, Ling Ren, Wenhua Cui","doi":"10.5755/j01.itc.52.2.32873","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32873","url":null,"abstract":"This paper puts forward a brain computer interface (BCI) system to realize the hand and wrist control using the ABB Mechanical Arm. This BCI system gathers four kinds of motor imaginary (MI) tasks (hand grasp, hand spread, wrist flexion and wrist extension) electroencephalogram (EEG) signals from 30 electrodes. It utilizes two fifth-order Butterworth Band-Pass Filter (BPF) with different bandwidths and normalization method to achieve the raw MI tasks EEG signals preprocessing. The main challenge of feature extraction is to extract enough representative features from MI tasks to classify them. This proposed BCI system extracts eleven kinds of features in time domain and time-frequency domain and uses mutual information method to reduce the large dimension of the extracted features. In addition, the BCI system applies a single convolutional layer Convolutional neural networks (CNN) with 30 filters to implement the quaternary classification of MI tasks. Compared with early researches, the classification accuracy of this BCI system is increased by about 35%. The actual mechanical arm grasping control experiments verifies that this BCI system has good adaptability.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"100 1","pages":"358-366"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74258814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-15DOI: 10.5755/j01.itc.52.2.32778
Suma Christal, Mary Sundararajan, Prabhjot Kaur, Anupama Kaushik
In the current state of medical research, the diagnosis of heart disease has become a challenging medical objective. This diagnosis is dependent on a thorough and accurate review of the detailed medical test results and medical background of the patient. With the aid of the internet of things (IoT) and the huge advancements in the field of deep learning, researchers aim to produce intelligent monitoring systems that assist physicians in both predicting and diagnosing disorders. In this context, this work proposes a novel prediction model based on deep learning and Internet-of-Medical-Things for the efficient and real-time diagnosis of heart disease. In this work, data from the Cleveland dataset is used for training the proposed model and further the data that is gathered from the sensors in the IoMT environment is used for testing the prediction capability of the model. Chaotic Harris Hawk optimization algorithm is employed for the feature extraction from the data and these extracted features are further passed on to the classification stage where Enhanced Convolutional Neural Networks are utilized to classify whether the patient is affected by heart disease or not. In order to evaluate the performance of the proposed model, it is compared with the Machine learning models such as Support Vector Machine with Ant Colony Optimization(SVM-ACO), Random Forest with Particle Swarm Optimization(RF-PSO), Naive Bayes with Harris Hawk Optimization(NB-HHO), K Nearest Neighbor with Spiral Optimization(KNN-SPO). Also, the proposed model is compared against deep learning architectures such as VGG-16, ResNet, AlexNet,ZFNet. Further, the proposed model also outperforms two existing works taken from the literature, Faster R-CNN-ALO, and MDCNN-AEHO, with a higher accuracy of 99.2%.
{"title":"Heart Diseases Diagnosis Using Chaotic Harris Hawk Optimization with E-CNN for IoMT Framework","authors":"Suma Christal, Mary Sundararajan, Prabhjot Kaur, Anupama Kaushik","doi":"10.5755/j01.itc.52.2.32778","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.32778","url":null,"abstract":"In the current state of medical research, the diagnosis of heart disease has become a challenging medical objective. This diagnosis is dependent on a thorough and accurate review of the detailed medical test results and medical background of the patient. With the aid of the internet of things (IoT) and the huge advancements in the field of deep learning, researchers aim to produce intelligent monitoring systems that assist physicians in both predicting and diagnosing disorders. In this context, this work proposes a novel prediction model based on deep learning and Internet-of-Medical-Things for the efficient and real-time diagnosis of heart disease. In this work, data from the Cleveland dataset is used for training the proposed model and further the data that is gathered from the sensors in the IoMT environment is used for testing the prediction capability of the model. Chaotic Harris Hawk optimization algorithm is employed for the feature extraction from the data and these extracted features are further passed on to the classification stage where Enhanced Convolutional Neural Networks are utilized to classify whether the patient is affected by heart disease or not. In order to evaluate the performance of the proposed model, it is compared with the Machine learning models such as Support Vector Machine with Ant Colony Optimization(SVM-ACO), Random Forest with Particle Swarm Optimization(RF-PSO), Naive Bayes with Harris Hawk Optimization(NB-HHO), K Nearest Neighbor with Spiral Optimization(KNN-SPO). Also, the proposed model is compared against deep learning architectures such as VGG-16, ResNet, AlexNet,ZFNet. Further, the proposed model also outperforms two existing works taken from the literature, Faster R-CNN-ALO, and MDCNN-AEHO, with a higher accuracy of 99.2%.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"239 2 1","pages":"500-514"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72954229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：