Article
Computer Science, Artificial Intelligence
Cuc Duong, Vethavikashini Chithrra Raghuram, Amos Lee, Rui Mao, Gianmarco Mengaldo, Erik Cambria
Summary: Wildfires pose significant threats to life, property, well-being, and the environment. Limited research has been done on wildfires compared to other climate-related hazards. This article presents data mining on public opinions about wildfires in Australia from 2014 to 2021, analyzing key aspects such as topic of concern, sentiment polarization, and perceived emotions. The study proposes a data filtering approach for emotion quantification, showing more accurate results compared to existing lexicon approaches. Additionally, trends in tweets reflect the real-life damage caused by wildfires, and people associate wildfires with the impacts of climate change.
COGNITIVE COMPUTATION
(2023)
Article
Economics
Huosong Xia, Wuyue An, Jiaze Li, Zuopeng (Justin) Zhang
Summary: This paper develops an outlier knowledge management framework based on complex adaptive system theory and information theory, and applies advanced natural language processing technology in the context of the COVID-19 pandemic. The study shows that microblog data exhibits heterogeneity in the context of COVID-19, and the extracted outlier knowledge is incorporated into the knowledge base for guiding extreme public health events.
SOCIO-ECONOMIC PLANNING SCIENCES
(2022)
Article
Computer Science, Hardware & Architecture
Shiyang Cao, Xiao Ma, Jiangfeng Zeng, Ming Yi
Summary: Financial sentiment analysis aims to improve decision-making for financial researchers by extracting public opinions about an institution. This paper proposes a novel financial sentiment classification framework (FSCN) that takes into account both fresh and hot opinions, capturing temporal information and interacting with user opinions for a comprehensive decision. Experimental results on a real-world dataset show that the framework achieves state-of-the-art results and is capable of providing reliable service in the financial system.
COMPUTERS & ELECTRICAL ENGINEERING
(2023)
Article
Computer Science, Interdisciplinary Applications
Frederick G. Conrad, Johann A. Gagnon-Bartsch, Robyn A. Ferg, Michael F. Schober, Josh Pasek, Elizabeth Hou
Summary: Researchers found reasonably high correlations between the sentiment of tweets containing the word "jobs" and survey-based measures of consumer confidence in 2008-2009, but this relationship was not observed after 2011, casting doubt on the effectiveness of using tweets as an alternative to survey responses.
SOCIAL SCIENCE COMPUTER REVIEW
(2021)
Article
Health Care Sciences & Services
Mingyun Gu, Haixiang Guo, Jun Zhuang
Summary: This study examines the impact of emergency events on emotions and behaviors in social media, with a focus on the 2019 Wuxi viaduct collapse accident in China. It found that social media rules were adhered to, with changes in user behavior and popular discussion topics influenced by focused news updates. Additionally, user sentiment changes were positively correlated with information released by personal-authentication accounts as the news of the collapse developed.
Article
Computer Science, Artificial Intelligence
Ying Lian, Xuefan Dong
Summary: This study proposes a new discrete opinion dynamics model based on the sentiment-opinion transformation mechanism to analyze unrelated discrete opinions from social media data. Simulation experiments are conducted to study the effects of different factors on sentiment and opinion dynamics.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Green & Sustainable Science & Technology
Kaushal Kumar Bhagat, Sanjaya Mishra, Alakh Dixit, Chun-Yen Chang
Summary: During the COVID-19 pandemic, public opinion about online learning was predominantly positive, with blogs being more positive and opinionated compared to news articles.
Article
Information Science & Library Science
Praveen Ranjan Srivastava, Prajwal Eachempati
Summary: This study analyzed nearly 18,000 tweets on Twitter regarding the CAA act introduced in India, finding that the majority of opinions were negative but overall neutral. Recommendations were provided on improving public perception about the scheme by interpreting the Act more effectively.
JOURNAL OF GLOBAL INFORMATION MANAGEMENT
(2021)
Article
Computer Science, Information Systems
Prajwal Eachempati, Praveen Ranjan Srivastava, Zuopeng Justin Zhang
Summary: This study examines the impact of the Coronavirus on society and businesses through a multi-channel social media approach, with country-wise analysis showing that public opinion is unanimously negative. Impact on businesses, society, and education has been highlighted to help formulate country-specific strategies.
ENTERPRISE INFORMATION SYSTEMS
(2021)
Article
Computer Science, Cybernetics
Chong Li, Yuling Qu, Xinping Zhu
Summary: This paper proposes a new asynchronous network model for dynamic sentiment analysis of online public opinions. It extends previous static models and provides a new way to extract opinion evolution patterns in complex environments. The applications of the proposed model offer new insights into the management of online public opinions.
Article
Computer Science, Artificial Intelligence
Muhammad Mujahid, Furqan Rustam, Fahad Alasim, Muhammad Abubakar Siddique, Imran Ashraf, Raheem Yar Khan
Summary: With the rise of social media platforms, sharing reviews has become a social norm in today's modern society. Restaurants use sentiment analysis to analyze customer feedback and improve the quality of their products or services. This study focuses on deep ensemble models and compares their performance with lexicon-based methods for sentiment classification. The results show that deep ensemble models outperform lexicon-based methods, with BiLSTM+GRU achieving the highest accuracy of 95.31% in three class problems. Topic modeling reveals the presence of negative sentiments towards Subway restaurants.
PEERJ COMPUTER SCIENCE
(2023)
Article
Food Science & Technology
Lei Xia, Bo Chen, Kyle Hunt, Jun Zhuang, Cen Song
Summary: This research investigates the intersection of food safety and online social networking, utilizing natural language processing techniques and social network analysis. The study finds that Internet users form modular communities with different topics and emotional states, focusing on a wide range of food safety topics and indicating an overall increase in food safety awareness.
Article
Public, Environmental & Occupational Health
Yunfeng Shang, Fangbin Qian, Nan Gao, Qin Yang, Yiting Guo, Yunpeng Sun
Summary: This study examines the impact of public health emergencies on the stock prices of insurance companies and finds that these emergencies have a positive and persistent effect on insurance companies' portfolios through investors' sentiment. However, the fear index triggered by public health emergencies is negatively associated with insurance stock portfolio returns, and insurers with smaller market capitalization are more strongly influenced by investors' sentiment.
FRONTIERS IN PUBLIC HEALTH
(2022)
Article
Computer Science, Information Systems
Bin Wang, Enhui Wang, Zikun Zhu, Yangyang Sun, Yaodong Tao, Wei Wang
Summary: There are two main issues in current sentiment analysis models in social sensor networks: firstly, most models only analyze sentiment within the text and overlook users, making it difficult to explain experimental results; secondly, few studies extract specific opinions from users, bringing challenges to analyzing opinion evolution.
INTERNATIONAL JOURNAL OF DISTRIBUTED SENSOR NETWORKS
(2021)
Article
Multidisciplinary Sciences
Scott A. Condie, Corrine M. Condie
Summary: Understanding the development and persistence of polarised opinions in social networks is a key challenge. External events can have a broad impact on polarisation of opinions, influencing populations synchronistically across a large proportion.
SCIENTIFIC REPORTS
(2021)