Article
Immunology
Liviu-Adrian Cotfas, Liliana Craciun, Camelia Delcea, Margareta Stela Florescu, Erik-Robert Kovacs, Anca Gabriela Molanescu, Mihai Orzan
Summary: This paper analyzes English tweets posted worldwide during two different time periods following the announcement of the Delta and Omicron variants of COVID-19. By using a language model, the study detects tweets expressing vaccine hesitancy and analyzes the reasons behind it. The findings show an increase in hesitant tweets from 4.31% during the Delta period to 11.22% during the Omicron period, accompanied by a decrease in the number of reasons for vaccine hesitancy. This raises concerns about the effectiveness of vaccination information campaigns.
Review
Computer Science, Information Systems
Vibhuti, Neeru Jindal, Harpreet Singh, Prashant Singh Rana
Summary: This paper reviews the essential requirements for an automatic system to monitor people wearing face masks and conducts a comprehensive study of available techniques and their performance analysis. The pros and cons of each method are discussed, along with the sources of datasets and required software. The paper also discusses the use cases, limitations, and observations for the system, and concludes with several directions for future research.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Public, Environmental & Occupational Health
Asmit Kumar Singh, Paras Mehan, Divyanshu Sharma, Rohan Pandey, Tavpritesh Sethi, Ponnurangam Kumaraguru
Summary: This study assessed the use and fit of face masks and social distancing in the United States, particularly during the Black Lives Matter protests, through analysis of social media images. The results showed a significant decrease in group posting after the implementation of stay-at-home laws and an increase in mask use in cities where mask mandates were in place. However, a high percentage of posts still showed disregard for guidelines.
JMIR PUBLIC HEALTH AND SURVEILLANCE
(2022)
Article
Computer Science, Information Systems
Jwen Fai Low, Benjamin C. M. Fung, Farkhund Iqbal
Summary: COVID-19 presents an opportunity to examine public acceptance of universal masking, a healthcare intervention that is unfamiliar to many in the Anglosphere. Through the analysis of over two million tweets, we observed the strategies used by supporters and opponents of masking, as well as the key themes driving the discussion. Initially, pro-mask tweets dominated Twitter, but they were later challenged by anti-mask tweets. Engagement metrics favored pro-mask tweets at first, but gradually shifted towards anti-mask tweets. Furthermore, our analysis suggests the possibility of platform owners suppressing certain aspects of the mask-wearing debate.
Article
Health Care Sciences & Services
Joanne Chen Lyu, Garving K. Luli
Summary: This study analyzed the topics emerging from public discussion about the CDC on Twitter related to COVID-19, identifying major themes such as COVID-19 death counts, opinions about CDC credibility, and CDC's guidelines. These topics were categorized into four overarching themes: knowing the virus and the situation, policy and government actions, response guidelines, and general opinion about credibility, providing valuable insights for improving communication between public health agencies and the public during a prolonged pandemic like COVID-19.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Chemistry, Analytical
Tahani Alqurashi
Summary: The coronavirus has disrupted people's lives, and Saudi Arabia took quick actions to suppress the virus. In the education sector, distance learning was implemented for the entire 2020 academic year. Analyzing tweets and using machine learning models, the study found that the Saudi public predominantly supported distance education during this period.
Article
Computer Science, Information Systems
Ruben Yanez Martinez, Guillermo Blanco, Analia Lourenco
Summary: The paper introduces new annotated corpora for stance detection on Spanish Twitter data, particularly Health-related tweets. The research aims to develop a manually annotated benchmark corpus for emotion recognition in social posts, evaluate the efficiency of semi-supervised models for extending the corpus, and describe the short text corpora using specialized topic modeling. The results demonstrate that the self-training method with SVM base estimator is effective in reducing annotation work and achieving high model performance.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Infectious Diseases
Samira Yousefinaghani, Rozita Dara, Samira Mubareka, Andrew Papadopoulos, Shayan Sharif
Summary: This study identified public sentiments and opinions towards COVID-19 vaccines on Twitter, showing a dominance of positive sentiments but active discussions on vaccine rejection and hesitancy. Different countries exhibited varying patterns. Additionally, the study found that vaccine opposition content came partly from Twitter bots or political activists, while support for vaccination originated from well-known individuals and organizations.
INTERNATIONAL JOURNAL OF INFECTIOUS DISEASES
(2021)
Article
Computer Science, Information Systems
Iram Javed, Muhammad Atif Butt, Samina Khalid, Tehmina Shehryar, Rashid Amin, Adeel Muzaffar Syed, Marium Sadiq
Summary: The coronavirus spreads through airborne droplets and causes respiratory infections such as sneezing, coughing, and pneumonia. To mitigate the spread, the World Health Organization recommends avoiding public interactions and following SOPs like wearing face masks and maintaining social distancing. However, enforcing these SOPs on a large scale remains challenging. A deep learning-based visual object detection network is proposed for face mask detection in public spots. A large-scale dataset and an end-to-end pipeline for real-time face mask detection and social distance measurement are presented to accelerate development. The pipeline improved accuracy by 5.3% compared to the baseline version.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Review
Biotechnology & Applied Microbiology
Mahesh Ganesapillai, Bidisha Mondal, Ishita Sarkar, Aritro Sinha, Saikat Sinha Ray, Young-Nam Kwon, Kazuho Nakamura, K. Govardhan
Summary: The threat of epidemic outbreaks is increasing due to the exponential growth of the global population and human mobility. Personal protective equipment, such as face masks, are crucial in protecting against viral infections. However, the uncontrolled manufacture and disposal techniques of face masks pose potential threats. Improper solid waste management contributes to viral propagation and increases biomedical waste. The chemical constituents in single-use face masks, including plasticisers and flame retardants, can lead to health issues. Despite extensive research, the efficacy and post-disposal impact of personal protective equipment are yet to be adequately explored. This review provides an overview of different forms of personal protective equipment, discusses proper waste management techniques, and examines the innovations and impact of face masks on human health and the environment. Strategies for safe and proper solid waste disposal are also discussed, along with the development of a 3D model of a face mask using computational fluid dynamics. The review concludes with possibilities for future advancements and promising research avenues in personal protective equipment.
ENVIRONMENTAL TECHNOLOGY & INNOVATION
(2022)
Article
Multidisciplinary Sciences
Elyssa M. Barrick, Mark A. Thornton, Diana Tamir
Summary: Research shows that in the presence of masks, people tend to rely more on visual cues from the eye area to judge others' emotional similarity, indicating a shift in how people process facial information.
Article
Computer Science, Hardware & Architecture
Celestine Iwendi, Senthilkumar Mohan, Suleman Khan, Ebuka Ibeke, Ali Ahmadian, Tiziana Ciano
Summary: This research uses Information Fusion to obtain real news and fake news data, and utilizes deep learning models to detect fake news related to COVID-19. The results show that our model has achieved good performance in terms of accuracy and precision, surpassing traditional machine learning algorithms.
COMPUTERS & ELECTRICAL ENGINEERING
(2022)
Article
Computer Science, Information Systems
Trong Hieu Luu, Phan Nguyen Ky Phuc, Zhiqiu Yu, Duy Dung Pham, Huu Trong Cao
Summary: This article discusses the negative impact of the COVID-19 pandemic on social life and emphasizes the importance of wearing face masks in preventing virus transmission. To enhance the effectiveness of mask usage, a method for detecting and warning individuals who do not wear or misuse face masks is proposed using deep learning techniques. The approach achieves an accuracy of over 95%.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Remote Sensing
Xiao Huang, Siqin Wang, Mengxi Zhang, Tao Hu, Alexander Hohl, Bing She, Xi Gong, Jianxin Li, Xiao Liu, Oliver Gruebner, Regina Liu, Xiao Li, Zhewei Liu, Xinyue Ye, Zhenlong Li
Summary: Social media data mining has played a significant role in COVID-19 research, contributing to early warning, monitoring human mobility, conveying information, analyzing public attitudes and emotions, identifying misinformation, and detecting incidents of hatred and violence. This review summarizes the progress of social media data mining studies in the context of COVID-19, provides essential features of publicly available COVID-19 related social media data archives, discusses challenges in social media analytics, and presents visions for future research directions.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Multidisciplinary Sciences
Jun Lang, Wesley W. Erickson, Zhuo Jing-Schmidt
Summary: The study found a clear rhetorical polarization between pro- and anti-mask hashtags among American Twitter users, with exponential frequency increases of both types of hashtags during the pandemic. Participation-wise, pro-mask hashtags dominated, creating an echo chamber effect that ignored the rhetoric of the anti-mask minority.
Article
Mathematics, Interdisciplinary Applications
Camelia Delcea, Liviu-Adrian Cotfas, R. John Milne, Naiming Xie, Rafal Mierzwiak
Summary: This paper investigates the back-to-front boarding method and its variations used during the COVID-19 outbreak, clustering the variations into three clusters using grey clustering to improve boarding efficiency and passenger comfort. The findings help airlines better understand how to select the boarding method that best fits their policies and goals under the new conditions imposed by the pandemic.
GREY SYSTEMS-THEORY AND APPLICATION
(2022)
Article
Transportation
R. John Milne, Liviu-Adrian Cotfas, Camelia Delcea
Summary: The Reverse Pyramid boarding method is more effective in reducing health risks, but the optimal performance is influenced by the volume of carry-on luggage, social distancing while walking, and the number of boarding groups.
TRANSPORTMETRICA B-TRANSPORT DYNAMICS
(2022)
Article
Computer Science, Artificial Intelligence
Liviu-Adrian Cotfas, Camelia Delcea, Simone Mancini, Cristina Ponsiglione, Luigi Vitiello
Summary: The cruise ship industry has witnessed a significant rise in passenger numbers over the years, with a growth of 60.11% from 2009 to 2018, reaching 28.5 million passengers. The importance of ensuring passenger safety during ship evacuation has been recognized, leading to continuous efforts by cruise ship companies to enhance their evacuation plans. The use of smartphone applications has emerged as an effective measure in improving evacuation time, as indicated by an agent-based model simulation conducted in this study. The findings demonstrate that the adoption of such applications can significantly reduce evacuation time.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Multidisciplinary Sciences
Marcel-Ioan Bolos, Ioana-Alexandra Bradea, Camelia Delcea
Summary: This paper proposes a model for the covariance of financial assets using neutrosophic fuzzy numbers. The concepts of neutrosophic covariance and independent neutrosophic portfolios are discussed and applied. A three-step approach is suggested to identify the return, risk, and structure of the independent neutrosophic portfolio. Neutrosophic fuzzy theory is chosen for its ability to properly model financial performance indicators, even when linguistic variables are used. Numerical examples are provided for better understanding, and the results can be used to support decisions made by capital market investors.
Article
Computer Science, Cybernetics
Camelia Delcea, Saad Ahmed Javed, Margareta-Stela Florescu, Corina Ioanas, Liviu-Adrian Cotfas
Summary: The Grey System Theory (GST) is an emerging area of research within artificial intelligence. This study collected and analyzed research papers utilizing GST in the fields of economics and education from the Web of Science database. The study identified prominent authors, institutions, publications, and journals associated with GST, as well as extracted and analyzed significant keywords, trends, and research directions. The analysis revealed a growing emphasis on the application of GST in diverse domains, including supply chain management, manufacturing, and economic development.
Article
Immunology
Liviu-Adrian Cotfas, Liliana Craciun, Camelia Delcea, Margareta Stela Florescu, Erik-Robert Kovacs, Anca Gabriela Molanescu, Mihai Orzan
Summary: This paper analyzes English tweets posted worldwide during two different time periods following the announcement of the Delta and Omicron variants of COVID-19. By using a language model, the study detects tweets expressing vaccine hesitancy and analyzes the reasons behind it. The findings show an increase in hesitant tweets from 4.31% during the Delta period to 11.22% during the Omicron period, accompanied by a decrease in the number of reasons for vaccine hesitancy. This raises concerns about the effectiveness of vaccination information campaigns.
Article
Green & Sustainable Science & Technology
Alexandra-Nicoleta Ciucu-Durnoi, Margareta Stela Florescu, Camelia Delcea
Summary: This paper aims to assess the extent to which Romania is projected to achieve its sustainable development goals. Using the ARIMA method to forecast values for a period of three years, the analysis found that Romania has made good progress in some objectives, but not in all indicators. It should be considered that Romania occupies low positions in terms of progress at the European level and some goals may not be met.
Article
Green & Sustainable Science & Technology
Stefan Ionescu, Nora Chirita, Ionut Nica, Camelia Delcea
Summary: Using machine learning techniques and agent-based modeling, this research explores the impact of connectivity and interconnection between commercial banks in Romania on the emergence and propagation of financial contagion effects.
Article
Computer Science, Information Systems
Erik-Robert Kovacs, Liviu-Adrian Cotfas, Camelia Delcea, Margareta-Stela Florescu
Summary: This paper uses a data-driven approach to analyze the evolution of opinions about COVID-19 vaccination on social media throughout the pandemic, based on the gender of the authors. The analysis shows that most tweets have a neutral stance, with more tweets in favor of vaccination compared to tweets opposing vaccination, although this distribution changes over time in response to specific events. The subject matter of the tweets varies more between stances than between genders.
Article
Chemistry, Multidisciplinary
Adrian Domenteanu, Camelia Delcea, Nora Chirita, Corina Ioanas
Summary: This paper presents a bibliometric analysis of the utilization of agent-based modeling in the field of transportation, revealing a consistent and robust growth in scholarly interest over the considered period. The study identifies key contributors, affiliations, influential publications, and renowned journals in this domain. The findings show a distinct upward trajectory of agent-based modeling in transportation since 2008, with a significant surge in paper production and prominent development in air transport and road transport domains.
APPLIED SCIENCES-BASEL
(2023)
Article
Multidisciplinary Sciences
R. John Milne, Liviu-Adrian Cotfas, Camelia Delcea, Liliana Craciun, Anca Gabriela Molanescu
Summary: This study investigates boarding policies adapted for the pandemic with varying levels of passenger compliance. The WilMA-Spread and Reverse-pyramid-Spread boarding methods are found to be effective. Non-compliance with the prescribed aisle social distance can impact health metrics, but it shortens the time required for boarding the airplane.
Article
Mathematics, Interdisciplinary Applications
Camelia Delcea, Liviu-Adrian Cotfas, Rafal Mierzwiak, Corina Ioanas
Summary: The COVID-19 pandemic has had a significant impact on the airline industry, resulting from reduced flights, protocols, restrictions, and passenger reluctance to travel by plane. In response, airlines have tried to create a safe boarding process with social distancing measures. Research shows that the Reverse Pyramid boarding method is more effective in terms of time and health risks compared to other methods. This paper aims to determine which variations of Reverse Pyramid can be used for boarding through the front door of an airplane, using an agent-based model and grey clustering to categorize and analyze performance based on boarding time, aisle seat risk, and window seat risk.
JOURNAL OF GREY SYSTEM
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Erik-Robert Kovacs, Liviu-Adrian Cotfas, Camelia Delcea
Summary: Social media platforms can be both beneficial for democratic debate and a source of unhealthy discourse. This study analyzes Twitter data to understand the evolution and attributes of unhealthy conversation. The results aim to improve understanding of the impact of unhealthy discourse on political polarization and partisanship.
ADVANCES IN COMPUTATIONAL COLLECTIVE INTELLIGENCE, ICCCI 2022
(2022)
Article
Computer Science, Information Systems
Liviu-Adrian Cotfas, Camelia Delcea, Livia-Diana Iancu, Corina Ioanas, Cristina Ponsiglione
Summary: This paper explores the usage of agent-based modeling in large event halls evacuation during music festivals and cultural events. It proposes an adapted cone exit approach to facilitate the guidance of the agents in the model and compares its advantages with the classical cone exit approach. Different evacuation scenarios are simulated and analyzed to observe the capabilities of evacuation modeling in emergency situations. The visual interface of the agent-based model allows for the identification of factors that may contribute to a prolonged evacuation process and potential measures to ensure a safe evacuation process.