Pub Date : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035601
Mike Louie C. Enriquez, Ronnie S. Concepcion, R. Relano, Kate G. Francisco, Adrian Genevie G. Janairo, Jonah Jahara G. Baun, J. A. D. Leon, R. R. Vicerra, A. Bandala, E. Dadios
Electromagnetic shielding has been a well-known mechanism to protect electronic circuits and antennas against unnecessary noise and signals in the system. Thus, this paper has introduced a shielding for very low-frequency subsurface imaging in the form of an antenna enclosure that will help in efficient power transfer and redirection of the signal. With that, the AHP technique was utilized for ranking the sub-criteria and alternatives concerning electrical qualities, material quality features, and manufacturing factors for selecting the best material for an antenna enclosure. The substrate materials which are also referred as alternatives include the Roger 5880 woven glass reinforced hydrocarbon/ceramics, FR4 glass-reinforced epoxy laminate material, and epoxy resin. Furthermore, an AHP calculator was utilized to perform the consistency ratio computations where its value must not be greater than 0.1. Based on the findings, the most suited material for a substrate material is epoxy resin dielectric substrate which considerably satisfies the electrical and mechanical requirements in the dipole plate of an underground imaging antenna. With a 0.373 weight value, it outperformed the Roger 5880 and the FR4 with 0.309 and 0.300 weight values, respectively, in total weight ranking alternatives. This claim was verified technically by performing electrical antenna simulations in Altair Feko on a one-pair equatorial dipole-dipole antenna. The epoxy resin practical reading of 6.354 μA, 7.567 kV/m, 8.580 kV/m, 87.651% 1.020 W of receiver current, inner and outer enclosure electric fields, power efficiency, and power loss, respectively, operated at 3 kHz which is ideal for long wave soil exploration.
{"title":"Analytical Hierarchy Process-based Selection of Underground Imaging Antenna Enclosure Material for Optimized Power Transfer","authors":"Mike Louie C. Enriquez, Ronnie S. Concepcion, R. Relano, Kate G. Francisco, Adrian Genevie G. Janairo, Jonah Jahara G. Baun, J. A. D. Leon, R. R. Vicerra, A. Bandala, E. Dadios","doi":"10.1109/IMCOM56909.2023.10035601","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035601","url":null,"abstract":"Electromagnetic shielding has been a well-known mechanism to protect electronic circuits and antennas against unnecessary noise and signals in the system. Thus, this paper has introduced a shielding for very low-frequency subsurface imaging in the form of an antenna enclosure that will help in efficient power transfer and redirection of the signal. With that, the AHP technique was utilized for ranking the sub-criteria and alternatives concerning electrical qualities, material quality features, and manufacturing factors for selecting the best material for an antenna enclosure. The substrate materials which are also referred as alternatives include the Roger 5880 woven glass reinforced hydrocarbon/ceramics, FR4 glass-reinforced epoxy laminate material, and epoxy resin. Furthermore, an AHP calculator was utilized to perform the consistency ratio computations where its value must not be greater than 0.1. Based on the findings, the most suited material for a substrate material is epoxy resin dielectric substrate which considerably satisfies the electrical and mechanical requirements in the dipole plate of an underground imaging antenna. With a 0.373 weight value, it outperformed the Roger 5880 and the FR4 with 0.309 and 0.300 weight values, respectively, in total weight ranking alternatives. This claim was verified technically by performing electrical antenna simulations in Altair Feko on a one-pair equatorial dipole-dipole antenna. The epoxy resin practical reading of 6.354 μA, 7.567 kV/m, 8.580 kV/m, 87.651% 1.020 W of receiver current, inner and outer enclosure electric fields, power efficiency, and power loss, respectively, operated at 3 kHz which is ideal for long wave soil exploration.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124008830","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035605
Yong Wang, Yaqi Li, Dingsheng Wang
As the extraction of small water bodies in remote sensing images has problems such as water line interruption and pretzel phenomenon, in order to be able to improve the extraction accuracy of small water bodies, this paper proposes a small water body extraction method based on Res2Net- Unet. The method uses the encoder and decoder structure of the UNet model. Firstly, the ResNet-50 network of the Res2Net module is used as an encoder, thus exploiting the feature information at multiple scales in the image. Secondly, a hybrid domain attention mechanism is incorporated into the decoder structure to fully mine the spatial and channel features in the image. Finally, a jump connection is added between the encoder and decoder to better fuse the features extracted by the encoder and decoder. Experiments on the Chinese Gaofen-1(GF-1) image datasets from two study areas show that the method in this paper is feasible for more complete and more accurate extraction of small water bodies compared with common deep learning models.
{"title":"Extraction of small water body information based on Res2Net-Unet","authors":"Yong Wang, Yaqi Li, Dingsheng Wang","doi":"10.1109/IMCOM56909.2023.10035605","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035605","url":null,"abstract":"As the extraction of small water bodies in remote sensing images has problems such as water line interruption and pretzel phenomenon, in order to be able to improve the extraction accuracy of small water bodies, this paper proposes a small water body extraction method based on Res2Net- Unet. The method uses the encoder and decoder structure of the UNet model. Firstly, the ResNet-50 network of the Res2Net module is used as an encoder, thus exploiting the feature information at multiple scales in the image. Secondly, a hybrid domain attention mechanism is incorporated into the decoder structure to fully mine the spatial and channel features in the image. Finally, a jump connection is added between the encoder and decoder to better fuse the features extracted by the encoder and decoder. Experiments on the Chinese Gaofen-1(GF-1) image datasets from two study areas show that the method in this paper is feasible for more complete and more accurate extraction of small water bodies compared with common deep learning models.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129554764","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035603
Bojun Wang, Sajid Ali, Xinyi Fan, Tamer Abuhmed
Cameras are becoming more pervasive and ubiquitous. The daily activities of individuals are being captured by millions of cameras in public spaces, while individuals are obtaining massive amounts of egocentric videos by employing wearable cameras intended for life-logging. However, recording devices are inexpensive, highly computational, and inconvenient for privacy. We used a low-resolution infrared sensor to detect human activity, including sitting, standing, and lying down, and to locate humans. We acquired the data from a low-cost infrared device and preprocessed them to train the YOLO-v5 network. We developed and tested an infrared technology-based system consisting of $32 times 24$ thermal input. Our proposed model is trained on 3,864 low-resolution images and made publicly available. The trained YOLO-v5 achieved 96.34% mean Average Precision (mAP) using our designed lightweight and low-cost activity recognition device. We proposed Artificial Intelligence of Things (A-IoT) system can be used either as a stand-alone data collection such as an IoT device or as a data processing and analysis sub-center. Our system consists of a low-power edge computing device and a cost-effective low-resolution infrared module. Our proposed dataset is now available at https://github.com/InfoLab-SKKU/Thermal-Human-Detection
{"title":"Real-time human detection and behavior recognition using low-cost hardware","authors":"Bojun Wang, Sajid Ali, Xinyi Fan, Tamer Abuhmed","doi":"10.1109/IMCOM56909.2023.10035603","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035603","url":null,"abstract":"Cameras are becoming more pervasive and ubiquitous. The daily activities of individuals are being captured by millions of cameras in public spaces, while individuals are obtaining massive amounts of egocentric videos by employing wearable cameras intended for life-logging. However, recording devices are inexpensive, highly computational, and inconvenient for privacy. We used a low-resolution infrared sensor to detect human activity, including sitting, standing, and lying down, and to locate humans. We acquired the data from a low-cost infrared device and preprocessed them to train the YOLO-v5 network. We developed and tested an infrared technology-based system consisting of $32 times 24$ thermal input. Our proposed model is trained on 3,864 low-resolution images and made publicly available. The trained YOLO-v5 achieved 96.34% mean Average Precision (mAP) using our designed lightweight and low-cost activity recognition device. We proposed Artificial Intelligence of Things (A-IoT) system can be used either as a stand-alone data collection such as an IoT device or as a data processing and analysis sub-center. Our system consists of a low-power edge computing device and a cost-effective low-resolution infrared module. Our proposed dataset is now available at https://github.com/InfoLab-SKKU/Thermal-Human-Detection","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127416598","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035569
Abdulaziz Aborujilah, J. Adamu, S. A. Mokhtar, AbdulAleem Al-Othmani, Elham Yahya Al-alwi, D. Al-Hidabi
In Malaysia, the early days of e-learning implementations can be traced back to the late 1990s. However, the development and use of such systems are fraught with security issues, concerns about the security of personal data, and content-related security flaws, confidentiality, integrity, availability, authentication, and deliberate attacks are the list of security issues that are identified in the e-learning systems. More attention has been given to course creation and delivery and technology adoption. However, user privacy and security in the context of e-learning security have not been studied well. In this paper, an in-depth review has been conducted regarding security threats and their remedies to the e-learning systems, which consequently lead Malaysia's higher education to take precautionary measures to facilitate learners with a secure learning environment.
{"title":"CIA-based Analysis for E-Leaming Systems Threats and Countermeasures in Malaysian Higher Education: Review Paper","authors":"Abdulaziz Aborujilah, J. Adamu, S. A. Mokhtar, AbdulAleem Al-Othmani, Elham Yahya Al-alwi, D. Al-Hidabi","doi":"10.1109/IMCOM56909.2023.10035569","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035569","url":null,"abstract":"In Malaysia, the early days of e-learning implementations can be traced back to the late 1990s. However, the development and use of such systems are fraught with security issues, concerns about the security of personal data, and content-related security flaws, confidentiality, integrity, availability, authentication, and deliberate attacks are the list of security issues that are identified in the e-learning systems. More attention has been given to course creation and delivery and technology adoption. However, user privacy and security in the context of e-learning security have not been studied well. In this paper, an in-depth review has been conducted regarding security threats and their remedies to the e-learning systems, which consequently lead Malaysia's higher education to take precautionary measures to facilitate learners with a secure learning environment.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"157 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126188354","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035641
Talha Nazar, Shujaat Hussain Kausar, Kifayat-Ullah Khan
In explaining complicated concepts, infographics are a far more effective medium of communication than regular prose. In recent years, deep learning has seen a lot of success in a range of applications requiring pattern identification and artificial intelligence. One of these applications is image recognition. There are many kinds of infographics that may be used in resumes and CVs to demonstrate the degree of competence. The objective was to identify those infographics that were already existing in the CV and to determine the word that corresponded to each infographic before attempting to measure each infographic using numeric characters. The YOLO algorithm was used to identify infographics, while OCR was utilized in order to identify associated words. The filled component of the infographic was distinguished from the unfilled area by using the image intensity histogram analysis, thresholding and contours.
{"title":"Localizing and Analyzing the Infographics in Document Using Deep Learning","authors":"Talha Nazar, Shujaat Hussain Kausar, Kifayat-Ullah Khan","doi":"10.1109/IMCOM56909.2023.10035641","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035641","url":null,"abstract":"In explaining complicated concepts, infographics are a far more effective medium of communication than regular prose. In recent years, deep learning has seen a lot of success in a range of applications requiring pattern identification and artificial intelligence. One of these applications is image recognition. There are many kinds of infographics that may be used in resumes and CVs to demonstrate the degree of competence. The objective was to identify those infographics that were already existing in the CV and to determine the word that corresponded to each infographic before attempting to measure each infographic using numeric characters. The YOLO algorithm was used to identify infographics, while OCR was utilized in order to identify associated words. The filled component of the infographic was distinguished from the unfilled area by using the image intensity histogram analysis, thresholding and contours.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122736816","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035637
A. Rodionov
Methods for preparing conditions for the efficient use of graph structure decomposition are proposed, which makes it possible to speed up the calculation of its general terminal reliability, that is, the probability that all its vertices will be connected in the case of reliable vertices and unreliable edges that can fail independently with given probabilities. The main idea is to enumerate the states of edges in an arbitrary section and use known equations for decomposition in the cases of 1, 2 or 3 reliable edges. A technique for finding a cut that divides the graph approximately in half is proposed. Experiments show that the proposed approach is quite effective.
{"title":"Little tricks leading to a significant acceleration of the calculation of the reliability of a random graph","authors":"A. Rodionov","doi":"10.1109/IMCOM56909.2023.10035637","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035637","url":null,"abstract":"Methods for preparing conditions for the efficient use of graph structure decomposition are proposed, which makes it possible to speed up the calculation of its general terminal reliability, that is, the probability that all its vertices will be connected in the case of reliable vertices and unreliable edges that can fail independently with given probabilities. The main idea is to enumerate the states of edges in an arbitrary section and use known equations for decomposition in the cases of 1, 2 or 3 reliable edges. A technique for finding a cut that divides the graph approximately in half is proposed. Experiments show that the proposed approach is quite effective.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128593090","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035594
Saki Inoue, Yuanyuan Wang, Yukiko Kawai, K. Sumiya
e-Learning has become a common practice in ed-ucational institutions, giving students more opportunities for remote learning. Learners generally need to spend some time formulating and thinking about their questions related to the learning material. Therefore, in this study, we concentrate on video lectures that individuals can access at their own convenience. We focus on critical thinking, which is defined as the process of critically examining the foundations of an idea or belief. We aim to provide learners with video lectures that aid in the growth and improvement of their critical thinking abilities. In order to do this, we propose an approach for helping and supporting questions generation as well as an approach for presenting relevant slides in the question support task. The findings of a questionnaire are used to evaluate the effectiveness of our proposed approaches. We completed evaluations three times utilizing lecture slides from various academic areas to verify the effectiveness of our proposed slide rankings. The findings show that based on the lecture's subject, different lecture slides are recommended for each question. On the other hand, our proposed approach tended to be highly evaluated for questions containing multiple keywords. Furthermore, the results indicate that the proposed slide rankings are more useful in difficult-to-understand slides.
{"title":"Learning Support System using Question Generation to Encourage Critical Thinking in Online Lectures","authors":"Saki Inoue, Yuanyuan Wang, Yukiko Kawai, K. Sumiya","doi":"10.1109/IMCOM56909.2023.10035594","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035594","url":null,"abstract":"e-Learning has become a common practice in ed-ucational institutions, giving students more opportunities for remote learning. Learners generally need to spend some time formulating and thinking about their questions related to the learning material. Therefore, in this study, we concentrate on video lectures that individuals can access at their own convenience. We focus on critical thinking, which is defined as the process of critically examining the foundations of an idea or belief. We aim to provide learners with video lectures that aid in the growth and improvement of their critical thinking abilities. In order to do this, we propose an approach for helping and supporting questions generation as well as an approach for presenting relevant slides in the question support task. The findings of a questionnaire are used to evaluate the effectiveness of our proposed approaches. We completed evaluations three times utilizing lecture slides from various academic areas to verify the effectiveness of our proposed slide rankings. The findings show that based on the lecture's subject, different lecture slides are recommended for each question. On the other hand, our proposed approach tended to be highly evaluated for questions containing multiple keywords. Furthermore, the results indicate that the proposed slide rankings are more useful in difficult-to-understand slides.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"14 7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125620343","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035653
Hana Munira Muhd Mukhtar, R. Ismail
A new technique of Blackboard architecture is presented herein, the implementation of a system that dynamically decides its action based on certain reasons that correspond to the problem-solving plan and strategies. We describe the principal elements of the blackboard; the design of domain and control knowledge sources (KSs). The mechanism of problem-solving uses a blackboard on how domain and control KSs integrate to solve the problem. The concept being used and the problem that they applied is supported by the trigger and feasible conditions as well as an adaptive scheduler that decides its own actions corresponding to the current interest of problem-solving at that particular time. The implementation of these new concepts on the blackboard is presented in timber harvesting decision-making for sustainable forest management.
{"title":"A New Technique of Blackboard Architecture in Decision-Making for Timber Harvesting","authors":"Hana Munira Muhd Mukhtar, R. Ismail","doi":"10.1109/IMCOM56909.2023.10035653","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035653","url":null,"abstract":"A new technique of Blackboard architecture is presented herein, the implementation of a system that dynamically decides its action based on certain reasons that correspond to the problem-solving plan and strategies. We describe the principal elements of the blackboard; the design of domain and control knowledge sources (KSs). The mechanism of problem-solving uses a blackboard on how domain and control KSs integrate to solve the problem. The concept being used and the problem that they applied is supported by the trigger and feasible conditions as well as an adaptive scheduler that decides its own actions corresponding to the current interest of problem-solving at that particular time. The implementation of these new concepts on the blackboard is presented in timber harvesting decision-making for sustainable forest management.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"126 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129785339","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035662
S. Wijaya, I. Handoko
This paper discusses the structure of digital conversations about Covid-19 on Twitter. Social network analysis was adopted to examine the relationship structure amongst Twitter users engaged within the network of conversations, and what kind of content users communicate with each other. We collected 97.067 tweets since March 2020 until April 2021, then analyzed the tweet conversations using NodeXL software. The results show that the network of conversations was a low-density network with a low reciprocity vertex pairs ratio. This suggests that the conversations were not effectively built. The top-ten most popular actors engaged in the conversations were dominated by government institution accounts. There were also many mentions to popular accounts. However, popular actors did not actively respond to the conversations that ensued. These findings emphasize that during the global health crisis, which was characterized by uncertain situations, people were inclined to search for related information on a daily basis from authorities. Some users also mentioned suggestions to the authorities to share official information related to the constantly changing situations. This study also highlighted that Twitter was able to facilitate important conversations, because of its capability to distribute information widely and quickly.
{"title":"Examining the Most Popular Users in Twitter Covid-19 Conversations","authors":"S. Wijaya, I. Handoko","doi":"10.1109/IMCOM56909.2023.10035662","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035662","url":null,"abstract":"This paper discusses the structure of digital conversations about Covid-19 on Twitter. Social network analysis was adopted to examine the relationship structure amongst Twitter users engaged within the network of conversations, and what kind of content users communicate with each other. We collected 97.067 tweets since March 2020 until April 2021, then analyzed the tweet conversations using NodeXL software. The results show that the network of conversations was a low-density network with a low reciprocity vertex pairs ratio. This suggests that the conversations were not effectively built. The top-ten most popular actors engaged in the conversations were dominated by government institution accounts. There were also many mentions to popular accounts. However, popular actors did not actively respond to the conversations that ensued. These findings emphasize that during the global health crisis, which was characterized by uncertain situations, people were inclined to search for related information on a daily basis from authorities. Some users also mentioned suggestions to the authorities to share official information related to the constantly changing situations. This study also highlighted that Twitter was able to facilitate important conversations, because of its capability to distribute information widely and quickly.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129863252","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 : 2023-01-03DOI: 10.1109/IMCOM56909.2023.10035647
Sammy Yap Xiang Bang, Kim-Ngoc Thi Le, D. Le, Hyunseung Choo
Retinal fundus diseases without immediate diagnoses and treatment may lead to serious consequences such as permanent visual impairment. Recently, many machine learning (ML) and deep learning (DL) models have been introduced for fundus image classification. However, those heavy models require high-end graphics processing units for training and testing, thus not suitable for real-case usage in fundus cameras with limited computation power. In this paper, we demonstrate the effectiveness of our proposed two-block model, Feature Exploitation Lightweight network (FEL-net) which consists of a Feature Exploitation Block (FEB) and Lightweight Classification Block (LCB) by comparing it with other DL models. The experiment was carried out on a dataset of 21, 697 fundus images and our model achieves 99% binary classification accuracy. The proposed robust FEB is used to generate a refined feature pool for fundus images to build an efficient ML classifier that can distinguish fundus images with disease from the normal case with high accuracy and low running time.
{"title":"Feature Pool Exploitation for Disease Detection in Fundus Images","authors":"Sammy Yap Xiang Bang, Kim-Ngoc Thi Le, D. Le, Hyunseung Choo","doi":"10.1109/IMCOM56909.2023.10035647","DOIUrl":"https://doi.org/10.1109/IMCOM56909.2023.10035647","url":null,"abstract":"Retinal fundus diseases without immediate diagnoses and treatment may lead to serious consequences such as permanent visual impairment. Recently, many machine learning (ML) and deep learning (DL) models have been introduced for fundus image classification. However, those heavy models require high-end graphics processing units for training and testing, thus not suitable for real-case usage in fundus cameras with limited computation power. In this paper, we demonstrate the effectiveness of our proposed two-block model, Feature Exploitation Lightweight network (FEL-net) which consists of a Feature Exploitation Block (FEB) and Lightweight Classification Block (LCB) by comparing it with other DL models. The experiment was carried out on a dataset of 21, 697 fundus images and our model achieves 99% binary classification accuracy. The proposed robust FEB is used to generate a refined feature pool for fundus images to build an efficient ML classifier that can distinguish fundus images with disease from the normal case with high accuracy and low running time.","PeriodicalId":230213,"journal":{"name":"2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133132818","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：