Changes In Sentiment Research Articles

Chinese Public Attitudes and Opinions on Health Policies During Public Health Emergencies: Sentiment and Topic Analysis.

By the end of 2021, the new wave of COVID-19 sparked by the Omicron variant spread rapidly due to its highly contagious nature, affecting more than 170 countries worldwide. Nucleic acid testing became the gold standard for diagnosing novel coronavirus infections. As of July 2022, numerous cities and regions in China have implemented regular nucleic acid testing policies, which have had a significant impact on socioeconomics and people's lives. This policy has garnered widespread attention on social media platforms. This study took the newly issued regular nucleic acid testing policy during the COVID-19 pandemic as an example to explore the sentiment responses and fluctuations of netizens toward new policies during public health emergencies. It aimed to propose strategies for managing public opinion on the internet and provide recommendations for policy making and public opinion control. We collected blog posts related to nucleic acid testing on Weibo from April 1, 2022, to July 31, 2022. We used the topic modeling technique latent Dirichlet allocation (LDA) to identify the most common topics posted by users. We used Bidirectional Encoder Representations from Transformers (BERT) to calculate the sentiment score of each post. We used an autoregressive integrated moving average (ARIMA) model to examine the relationship between sentiment scores and changes over time. We compared the differences in sentiment scores across various topics, as well as the changes in sentiment before and after the announcement of the nucleic acid price reduction policy (May 22) and the lifting of the lockdown policy in Shanghai (June 1). We collected a total of 463,566 Weibo posts, with an average of 3799.72 (SD 1296.06) posts published daily. The LDA topic extraction identified 8 topics, with the most numerous being the Shanghai outbreak, nucleic acid testing price, and transportation. The average sentiment score of the posts was 0.64 (SD 0.31), indicating a predominance of positive sentiment. For all topics, posts with positive sentiment consistently outnumbered those with negative sentiment (χ27=24,844.4, P<.001). The sentiment scores of posts related to "nucleic acid testing price" decreased after May 22 compared with before (t120=3.882, P<.001). Similarly, the sentiment scores of posts related to the "Shanghai outbreak" decreased after June 1 compared with before (t120=11.943, P<.001). During public health emergencies, the topics of public concern were diverse. Public sentiment toward the regular nucleic acid testing policy was generally positive, but fluctuations occurred following the announcement of key policies. To understand the primary concerns of the public, the government needs to monitor social media posts by citizens. By promptly sharing information on media platforms and engaging in effective communication, the government can bridge the information gap between the public and government agencies, fostering a positive public opinion environment.

An Analysis of MUJIs Successful Strategy in the Chinese Market

Abstract: The Chinese consumer market has changed significantly in the last few years due to a tendency toward simplicity and a change in customer sentiments. MUJI has effectively entered the Chinese market by using its sustainable development idea and minimalist style, allowing it to adapt to this shift. This paper aims to examine the innovative marketing strategy employed by MUJI in the Chinese market, its capacity to achieve long-term growth, and its role as a symbol of the minimalist lifestyle, while demonstrating an ability to adapt to the changing needs of its customers. In this paper, both literature review and data analysis are employed to provide an in-depth study of MUJIs strategy to enter the Chinese market. The results show that Chinese consumers preference for MUJI products not only stems from their simple and practical design, but also coincides with the concept of minimalist aesthetics in traditional Chinese culture. In addition, MUJI's unique advantages in product quality, brand image and marketing strategy further enhance its competitiveness in the Chinese market. Through specific analysis, it comprehensively discusses the success factors of this Japanese brands minimalist design style in the Chinese market, which provides a reference for the design and marketing of other international brands in the Chinese market.

Understanding the Incongruent Brand Personalities on Social Media: Evidence from Indian Brands

Abstract Although the social media websites decreased the firms’ control over their brand personalities, there is little research on the incongruence between communicated and perceived brand personality of Indian service brands. The impact of bots and suspicious accounts on the online brand personalities also remained unexplored. We analyzed 58,024 tweets of nine Indian service brands using machine learning. Some of the newer brands are savvier in managing their brand personalities. The public sector brands have poorer alignment between communicated and perceived brand personalities. A counterintuitive finding is that the brand sentiment changes in value, but not in category, after removing the suspicious accounts.

Perceived Responses of International Tourists to Transportation and Tourism Services During Typhoons Faxai and Hagibis in Japan

There is a limited understanding on the information-seeking behavior of international tourists during disaster response scenarios due to the lack of empirical studies on crisis communication in Japan. This study clarifies the topics generated from both international tourists and official Twitter accounts by applying the embedding Bidirectional Encoder Representations from Transformers (BERT) topic model and examines the temporal sentiment changes toward transportation and tourism using the sentiment scores obtained from topic-based Valence Aware Dictionary and sEntiment Reasoner (VADER) sentiment analysis during disasters. A case study was conducted using Twitter data on Typhoons Faxai and Hagibis, which struck Japan in 2019. This study found differences in the topics generated among international tourists and officials in response and a continuous negative sentiment toward specific transportation services. The managerial implications of these findings regarding the use of social media in crisis communication in tourism are also discussed.

Application and Empirical Analysis of Random Volatility Model in Financial Markets

The Stochastic Volatility Model (SVM), as a key innovation in modern financial engineering, has profoundly changed our understanding of the volatility characteristics of financial markets. This model not only considers the volatility of asset prices as a time-varying random variable, but also simulates this uncertainty by introducing stochastic processes such as Brownian motion or more complex stochastic differential equations, thereby achieving a precise characterization of the dynamic characteristics of volatility. Compared to traditional fixed or historical average volatility assumptions, SVM can capture sudden changes in market sentiment, the impact of news events, and the influence of macroeconomic factors on volatility, providing investors with a more realistic perspective on the market. This article not only deeply analyzes the theoretical basis and construction methods of SVM, but also verifies its effectiveness and applicability in different financial market environments through empirical analysis and actual market data, providing valuable references and insights for financial practitioners, scholars, and policy makers.

Aspect based sentiment analysis of Twitter mobile phone reviews using LSTM and Convolutional Neural Network

The proliferation of online shopping has led to a surge in product reviews, providing valuable information to consumers. However, the overwhelming volume and subjective nature of these reviews make it difficult to assess product performance accurately. We propose a machine learning-based system that extracts sentiment from online reviews to address this challenge. Our system effectively identifies positive, negative, and neutral sentiments and classifies sentiment for specific product aspects. By offering concise and clear information, our system empowers consumers to make informed purchasing decisions and assists manufacturers in improving their products. Our proposed LSTMCNN model, trained on a dataset of 62,563 reviews, achieved impressive results with an accuracy of 95.84%, precision of 95.6% and recall of 95.8%. This significantly outperforms existing models, demonstrating the effectiveness of our approach. Moving forward, we aim to further enhance the accuracy of our system, track sentiment changes over time, and develop personalized product recommendations. These advancements will continue to increase the value and utility of online reviews in e-commerce.

Sentiment Analysis on Twitter Social Media Regarding Covid-19 Vaccination with Naive Bayes Classifier (NBC) and Bidirectional Encoder Representations from Transformers (BERT)

Abstract. The Covid-19 vaccine is an important tool to stop the Covid-19 pandemic, however, there are pros and cons from the public regarding this Covid-19 vaccine. Purpose: These responses were conveyed by the public in many ways, one of which is through social media such as Twitter. Responses given by the public regarding the Covid-19 vaccination can be analyzed and categorized into responses with positive, neutral or negative sentiments. Methods: In this study, sentiment analysis was carried out regarding Covid-19 vaccination originating from Twitter using the Naïve Bayes Classifier (NBC) and Bidirectional Encoder Representations from Transformers (BERT) algorithms. The data used in this study is public tweet data regarding the Covid-19 vaccination with a total of 29,447 tweet data in English. Result: Sentiment analysis begins with data preprocessing on the dataset used for data normalization and data cleaning before classification. Then word vectorization was performed with TF-IDF and data classification was performed using the Naïve Bayes Classifier (NBC) and Bidirectional Encoder Representations from Transformers (BERT) algorithms. From the classification results, an accuracy value of 73% was obtained for the Naïve Bayes Classifier (NBC) algorithm and 83% for the Bidirectional Encoder Representations from Transformers (BERT) algorithm. Novelty: A direct comparison between classical models such as NBC and modern deep learning models such as BERT offers new insights into the advantages and disadvantages of both approaches in processing Twitter data. Additionally, this study proposes temporal sentiment analysis, which allows evaluating changes in public sentiment regarding vaccination over time. Another innovation is the implementation of a hybrid approach to data cleansing that combines traditional methods with the natural language processing capabilities of BERT, which more effectively addresses typical Twitter data issues such as slang and spelling errors. Finally, this research also expands sentiment classification to be multi-label, identifying more specific sentiment categories such as trust, fear, or doubt, which provides a deeper understanding of public opinion.

A Systematic Review on Impact of Political Crisis and Election on Capital Market Returns

The capital markets worldwide are vastly impacted by market sentiments and investors tend to react and rebalance their portfolios time to time due to such changes in market sentiments from political events. This systematic review examines the impact of political crisis and election on the capital market returns worldwide. It employs a systematic review research method and has gone through 4573 articles, out of which 62 articles published in reputed journals between 1970 and 2023 were considered for review. Of the methodology of the studies reviewed, 43.5% is quantitative, 27.4% is mixed, 17.7% is qualitative, and the rest is conceptual. Researches published in the context of Asian countries contributed 40% of this reviewed study, followed by Europe, Africa, and North America, each with 10%, South America with 3%, Australia with 1%, and allied countries within unions or agreements with 26%. The results show that political crises significantly impact capital market returns in Asia. Studies in North American context indicate that before elections and during political crises, the capital market volatility significantly rises with decreased capital market returns. In Australia, political events cause financial market fluctuation. In Africa, political instability hinders long-term capital accumulation. Additionally, informal political instability in South America directly affects economic growth, while formal instability indirectly affects economic vulnerability. The outcomes of such research hold practical implications for policymakers in formulating portfolio strategies and, therefore, delineate avenues for research exploring the integrated effects of political events on capital market returns by amalgamating trading information from stock exchanges worldwide.

Fiscal policy in the Bundestag: Textual analysis and macroeconomic effects

Fiscal policy is made in parliaments. We go to the roots of changes of fiscal policy in Germany and use a novel data set on all parliamentary speeches in the Bundestag from 1960 to 2021. We propose an embedding-based approach, which allows the representation of words and documents in a shared vector space, in order to measure fiscal policy-related sentiment in parliamentary debates at a scale from contractionary to expansionary. We also distinguish between sentiment related to exogenous and endogenous fiscal policy. We put fiscal sentiment into a series of recursively-identified vector autoregressive models to show that a change in fiscal sentiment causes a shift in government spending and has significant effects on the macroeconomy. The results support the notion that the debate in parliament contains information for the identification of government spending shocks.

The investment of renewable energy: Is green bond a safe-haven to hedge U.S. monetary policy uncertainty?

This paper uses the bootstrap rolling-window Granger causality test to investigate the relationship between U.S. monetary policy uncertainty and green bond. The method addresses the limitations of ignoring the instability of coefficients and the time-varying relationship between the variables in previous literature. The results find both insignificant and significant relationship in the full period. One explanation of the insignificant relationship is that the changes of investor sentiment influence their judgment regarding to the monetary policy uncertainty and green bond, and thus the significance of the relationship between the variables. Besides, this paper finds the significant inter-relationship is time-varying. Specifically, in the sub-period when the market sentiment is low, the rising monetary policy uncertainty influences green bond price negatively. Meanwhile, the unexpected changes of green bond may also have both positive and negative effects on monetary policy uncertainty in different sub-periods. Thus, this paper proposes that the impact mechanism between the variables is effective when the investors are irrational. Furthermore, the findings may provide valuable implications for investors and governments, including adjusting investment strategy when the monetary policy is unclear, increasing monetary policy transparency, and monitoring speculative activity in green bond market. In addition, it is necessary to emphasize that the validity of the findings is limited to the U.S. and the period when investors’ sentiment experience large fluctuations. Finally, future research may consider expand the analysis to the green stock, as it is another important channel for renewable energy investment.

Interaction between investor sentiment, limits to arbitrage and the returns of stock market anomalies: evidence from the UK stock market

This study investigates the role of two prominent concepts in finance: limits to arbitrage and investor sentiment in stock prices. The study examines how changes in market-wide investor sentiment and limits to arbitrage can affect the performance of nine UK stock market anomalies. The extant literature relating to investor sentiment focuses mainly on the US stock market, whilst research on the UK market typically examines aggregated index-level data. In addition, previous studies have focused on examining investor sentiment and limits to arbitrage separately. Using data from UK-listed companies over the period January 1997 to December 2019, the study finds that five stock market anomalies were related to changes in UK investor sentiment and produced significantly higher returns following periods of high investor sentiment, while the effect of limits to arbitrage was mostly limited. However, the interaction analysis provided support to the limits to arbitrage theory and demonstrated that the effect of high investor sentiment on stock market anomalies was more pronounced when combined with high limits to arbitrage and had less effect during periods characterised by low limits to arbitrage.

The Combined Effects of Financial and Market Indicators on the Valuation of New Listings: Evidence from the Biopharmaceutical Sector

This study confirms that financial and market data have a significant impact on the valuation of newly listed companies, especially in the biopharmaceutical industry. Using multiple linear regression models, this paper found significant differences in the impact of EBITGR and EBITDAGR on price-to-earnings ratio (P/E Ratio) across years. Other indicators such as Total Asset Turnover, Return on Assets (ROA) and Debt to Equity Ratio (D/E Ratio) have weaker explanatory power and unstable influence. In addition, fluctuations in market valuations of listed companies are influenced by market sentiment and the macroeconomic policy environment. In high-growth, high-risk industries such as biopharmaceuticals, regulatory policies and listing structure reforms in the market can cause structural changes in model performance. At the same time, changes in market sentiment and regulation can also affect the explanatory power of financial indicators to some extent. Further analyses showed that market preferences for valuation changed over time during the study period, especially when the market moved significantly, driven by regulatory policy and the external environment. This makes the conclusions of the study's findings subject to uncertainty in different market environments. Overall, this study reveals a number of key factors affecting the market valuation of newly listed companies and highlights the importance of the interaction between market data and financial indicators. In future research, more attention should be paid to the role of different time periods, market conditions and industry characteristics to avoid potential valuation traps and provide more informative support to market investors and managers.

The Combined Effects of Financial and Market Indicators on the Valuation of New Listings: Evidence from the Biopharmaceutical Sector

This study confirms that financial and market data have a significant impact on the valuation of newly listed companies, especially in the biopharmaceutical industry. Using multiple linear regression models, this paper found significant differences in the impact of EBITGR and EBITDAGR on price-to-earnings ratio (P/E Ratio) across years. Other indicators such as Total Asset Turnover, Return on Assets (ROA) and Debt to Equity Ratio (D/E Ratio) have weaker explanatory power and unstable influence. In addition, fluctuations in market valuations of listed companies are influenced by market sentiment and the macroeconomic policy environment. In high-growth, high-risk industries such as biopharmaceuticals, regulatory policies and listing structure reforms in the market can cause structural changes in model performance. At the same time, changes in market sentiment and regulation can also affect the explanatory power of financial indicators to some extent. Further analyses showed that market preferences for valuation changed over time during the study period, especially when the market moved significantly, driven by regulatory policy and the external environment. This makes the conclusions of the study's findings subject to uncertainty in different market environments. Overall, this study reveals a number of key factors affecting the market valuation of newly listed companies and highlights the importance of the interaction between market data and financial indicators. In future research, more attention should be paid to the role of different time periods, market conditions and industry characteristics to avoid potential valuation traps and provide more informative support to market investors and managers.

The role of uncertainty and sentiment for intraday volatility connectedness between oil and financial markets

We quantify intraday volatility connectedness between oil and key financial assets and assess how it is related to uncertainty and sentiment measures. For that purpose, we integrate the well-known spillover methodology with a TVP VAR model estimated on a unique, vast dataset of roughly 300 thousand 5 min quotations for most heavily traded financial assets: crude oil, the US dollar, S&P 500 index, gold and US treasury bonds. This distinguishes our investigation from previous studies, which usually employ relatively short samples of daily or weekly data and focus on connectedness between two asset classes. We contribute to the literature across three margins. First, we document that market connectedness at intraday frequency presents a different picture on markets co-movement compared to the estimates obtained using daily data. Second, we show that at 5 min frequency volatility is mostly transmitted from the stock market and absorbed by the bond and dollar markets, with oil and gold markets being occasionally important for volatility transmission. Third, we present evidence that daily averages of intraday connectedness measures respond to changes in sentiment and market-specific uncertainty. Interestingly, our results contrast with earlier findings, as they show that connectedness among markets decreases in periods of high volatility owing to market-specific factors. Our study points to the importance of using high-frequency data in order to better understand financial and commodity markets dynamics.

Predicting Customer Sentiment in Social Media Interactions: Analyzing Amazon Help Twitter Conversations Using Machine Learning

Social media platforms, particularly Twitter, have become essential sources of data for various applications, including marketing and customer service. This study focuses on analyzing customer interactions with Amazon's official support account, "@AmazonHelp," to understand and predict changes in customer sentiment during these interactions. Using the Twitter API, we extracted English-language tweets mentioning "@AmazonHelp," pre-processed the data, and categorized conversations to facilitate analysis. The primary objectives were to classify changes in customer sentiment and predict the overall sentiment change based on initial sentiment. We conducted experiments using multiple machines learning algorithms, including K-nearest neighbor, Naive Bayes, Artificial Neural Network, Bayes Net, Support Vector Machine, Logistic Regression, and Bagging with RepTree. Our dataset comprised over 6,500 conversations, filtered to include those with four or more tweets. Results indicated that K-nearest neighbor and Support Vector Machine offered the best balance between accuracy and F-measure, while Bagging with RepTree achieved the highest accuracy but had a lower F-measure. This study demonstrates the potential of integrating sentiment analysis and machine learning to effectively predict customer sentiment in social networks, providing valuable insights for improving customer engagement strategies.

AI approach on identifying change in public sentiment for major events: Dubai Expo 2020

Existing research in identifying changes in public sentiment for major events faces limitations in capturing the intricacies of temporal changes, especially with large-scale data and diverse linguistic nuances. This study addresses this problem, with a particular focus on the globally acclaimed Dubai Expo 2020. The methodology employed in this research introduces a groundbreaking approach by leveraging advanced data engineering techniques on a massive corpus of 118,025 Tweets obtained from diverse users across 60 languages, covering the period from April 2021 to March 2023. The study pioneers the application of artificial intelligence (AI), natural language processing (NLP), and Large Language Models (LLMs) for sentiment analysis and topic modeling on Twitter discourse. Through the meticulous data engineering process, including language detection, dynamic translation, and sentiment analysis, the research identifies subtle yet statistically significant changes in public sentiment, as evidenced by ANOVA testing (p=0.018 in average positive sentiment, p=0.004 in average neutral sentiment, and p=0.005 in average negative sentiments). Additionally, the study innovatively extracts and analyzes support-related Tweets, revealing distinct phases in temporal domains (pre-Expo, Expo, Post-Expo1, and Post-Expo2) and yielding 11,116 support-related tweets. The application of NLP techniques further uncovers 19 topics from UAE-related Tweets, providing a comprehensive understanding of the dynamic landscape of public sentiment over a two-year period. This research contributes significantly to the field by offering a novel and comprehensive framework for analyzing public sentiment, particularly in the context of major events, and sheds light on its broader implications for event management and public perception analysis.

SNSMiner_VAC: Analyzing vaccination based on social network service data for safety surveillance

COVID-19 has created a major issue around the world, posing a serious threat to socioeconomic and human health, and there is still fear of new infectious diseases like this. In Korea, in response to COVID-19, the COVID-19 vaccine approved by the Ministry of Food and Drug Safety was sequentially administered to each group, raising national interest in the safety, side effects, and effectiveness of the vaccine, and securing the basis for related policies. Many attempts have been made for this. We are interested in a methodology to establish a scientific basis for how to respond to issues regarding vaccine stability due to the emergence of new infectious diseases. In this study, we use comments posted on Naver, the largest social platform in Korea, to obtain consumer sentiment information about vaccines and understand the change in Korean public sentiment in the first half of 2021 for each vaccine based on the analysis pipeline (SNSMiner_VAC). We investigated the Korean public’s response to the COVID-19 vaccination on social media from November 30, 2021, to January 31, 2022. We collected comments related to the COVID-19 vaccination using the Korean words for “COVID-19 vaccination” as keywords. Of the 329,559 comments, 240,322 were analyzed after preprocessing. We developed three lexicons (stopwords, sentiment words, and custom words related to COVID-19) for Korean natural language processing. Furthermore, we extracted five clusters — including symptoms, side effects, vaccine hesitancy, pro-vaccination, and infectious disease control policy — through network analysis. The sentiment analysis revealed that negative public opinion was prominent even after the commencement of the phased recovery of daily life. From these results, we were able to automatically collect data-based information on patients’ symptoms and side effects for the new vaccine and prepare and analyze the basis for collecting public opinions about vaccination. We hope that the pipeline proposed in this study can be used as a tool to create a data source for effective management policies for vaccines being developed against the emergence of new infectious diseases.

Business Confidence in Indonesia: Which Macroeconomic Factors Have Long-Term Impact?

Business confidence refers to the level of optimism or pessimism that business owners have about the prospects of their companies and the overall economy. Thus, the focus of this study is to examine the long-term impact of various macroeconomic factors—economic growth, government expenditure, interest rates, inflation, exchange rates, and the composite stock price index—on the business confidence index in Indonesia by utilizing monthly data from January 2009 to December 2022. We employ Dynamic Ordinary Least Squares (DOLS) and Fully-Modified Ordinary Least Squares (FMOLS) as the main methods, with Canonical Cointegrating Regressions (CCR) as a robustness check method. The study also utilizes pairwise Granger causality tests for a comprehensive analysis. The findings indicate that all macroeconomic factors significantly impact the business confidence index in the long term across all methodologies. Specifically, economic growth, inflation, and the composite stock price index exert a positive impact, while government expenditure, interest rates, and exchange rates indicate a negative impact on the business confidence index. This evidence emphasizes the importance for businesses to diligently monitor macroeconomic trends and understand the patterns in these indicators so that companies can better anticipate changes in business sentiment. Taking a long-term perspective when making strategic decisions and investments is also advisable, recognizing that the influence of macroeconomic factors on business confidence may be more pronounced over time.

Dynamics of Indonesian Public Opinion on the Rohingya Crisis in Time Perspective Using Traditional Machine Learning and Deep Learning

The Rohingya are an ethnic minority who currently still face persecution and discrimination in Myanmar, so they have to flee to neighboring countries, such as Indonesia. However, the polemic regarding the issue of the existence of Rohingya refugees in Indonesia still shows that there are differences of opinion between groups who support and oppose it. For this reason, this research aims to determine the dynamics of Indonesian public opinion regarding the Rohingya from 2015-2023 via Twitter, as well as find out the topics that are often discussed each year using LDA. This research compares classification methods using traditional machine learning algorithms (NB, SVM, LR, and DT) and deep learning algorithms (LSTM, GRU, LSTM-GRU, and GRU-LSTM). The research results show that the traditional machine learning algorithm, LR, has the highest accuracy. There has been a change in sentiment from initially being dominated by positive sentiment to negative sentiment which is more dominant in the last five years. The topics that are often discussed for positive sentiment are the support of the Indonesian people for the Rohingya in providing assistance and shelter, while the negative topics are related to concerns about the social, economic, and security impacts that may be caused by the presence of Rohingya refugees.

Sentiment and thematic analysis of faculty responses: Transition to online learning

This project aims to understand physics faculty responses to transitioning to online teaching during the COVID-19 pandemic. We surveyed 662 physics faculty from the United States following the Spring 2020 term; of these, 258 completed a follow-up survey after the Fall 2020 term. We used natural language processing to measure the sentiment scores of 364 Spring 2020 responses and another 134 Fall 2020 responses of physics faculty who completed an optional written prompt. Additionally, we determined the change in sentiment scores of the 100 individuals who responded to both surveys. These sentiment scores measured between −1 and 1 for completely negative and completely positive, respectively. Sentiment scores after Spring 2020 were slightly positive with a median value of 0.2347. The distribution of sentiment changes was approximately normally distributed with a mean centered near zero. Analysis suggests the average sentiment did not change from the initial to follow-up surveys. To identify major topics within the responses for both surveys, latent Dirichlet allocation analysis was applied to the data. The topic distribution for the initial survey is given as course modifications and technology, negative aspects of the transition—primarily with labs and cheating, exam and evaluation difficulties, and difficulties with student understanding. The topics were noticeably different in the follow-up survey with differences between Fall and Spring, cooperative learning strategies, strategies that worked in the remote space, and benefits of in-person labs. Published by the American Physical Society 2024