Proposed information retrieval systems using LDA topic modeling for answer finding of COVID 19 pandemic: A brief survey of approaches and techniques

Understanding public responses to emergencies, including outbreaks of diseases, is necessary and significant. A demonstration of how to separate papers about the virus Covid-19 into different topics using topic modeling techniques in several studies is introduced in this research article. Inthe field of machine learning, topic modeling is a major topic. Though primarily, it is used to build models. It provides a quick and easy way to mine data from unstructured textual data, with samples representing documents.The most extensively utilized subject modeling approaches are Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation (LDA). On the other hand, model creation can be tedious and repetitious, requiring costly and methodical sensitivity analyses to determine the ideal collection of model parameters. Moreover, comparing models frequently require time-consuming subjective opinions. The topic models assign a probability to each word in the vocabulary corpus related to one or more themes (LSA, LDA). Several LDA and LSA models with varied degrees of coherence were generated, and the model with the greatest degree of coherence was selected. This experiment demonstrates that LDA outperforms LSA. © 2023 Author(s).

Digital transformation in tourism: bibliometric literature review based on machine learning approach

Purpose: This bibliometric study provides an overview of research related to digital transformation (DT) in the tourism industry from 2013 to 2022. The goals of the research are as follows: (1) to identify the development of academic papers related to DT in the tourism industry, (2) to analyze dominant research topics and the development of research interest and research impact over time and (3) to analyze the change in research topics during the pandemic. Design/methodology/approach: In this study, the authors processed 3,683 papers retrieved from the Web of Science and Scopus. The authors performed different types of bibliometric analyses to identify the development of papers related to DT in the tourism industry. To reveal latent topics, the authors implemented topic modeling based on latent Dirichlet allocation with Gibbs sampling. Findings: The authors identified eight topics related to DT in the tourism industry: City and urban planning, Social media, Data analytics, Sustainable and economic development, Technology-based experience and interaction, Cultural heritage, Digital destination marketing and Smart tourism management. The authors also identified seven topics related to DT in the tourism industry during the Covid-19 pandemic;the largest ones are smart analytics, marketing strategies and sustainability. Originality/value: To identify research topics and their development over time, the authors applied a novel methodological approach – a smart literature review. This machine learning approach is able to analyze a huge amount of documents. At the same time, it can also identify topics that would remain unrevealed by a standard bibliometric analysis. © 2023, Peter Madzík, Lukáš Falát, Lukáš Copuš and Marco Valeri.

Computational vs. qualitative: analyzing different approaches in identifying networked frames during the Covid-19 crisis

Despite the increasing adaption of automated text analysis in communication studies, its strengths and weaknesses in framing analysis are so far unknown. Fewer efforts have been made to automatic detection of networked frames. Drawing on the recent developments in this field, we harness a comparative exploration, using Latent Dirichlet Allocation (LDA) and a human-driven qualitative coding process on three different samples. Samples were comprised of a dataset of 4,165,177 million tweets collected from Iranian Twittersphere during the Coronavirus crisis, from 21 January, 2020 to 29 April, 2020. Findings showed that while LDA is reliable in identifying the most prominent networked frames, it misses to detects less dominant frames. Our investigation also confirmed that LDA works better on larger datasets and lexical semantics. Finally, we argued that LDA could give us some primary intuitions, but qualitative interpretations are indispensable for understanding the deeper layers of meaning. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

Analysis of Topic and Sentiment Trends in Customer Reviews before and after Covid-19 Pandemic

The Covid-19 pandemic has impacted many people's lives. Many researches have studied the impact of the pandemic on customer opinion change regarding services, yet there are still few researches regarding the change towards products. As a product category that experienced a significant increase in sales since the pandemic began, headphones have become a suitable product category to analyze the change. To analyze the change, this paper aims to discover the topics that customers discuss in their reviews. Latent Dirichlet Allocation (LDA) is selected as the topic modeling method to obtain the topics (i.e., aspects of a product) that are discussed in the customer reviews. In the case study, six topics that are discussed by customers are discovered, i.e., Durability Issues, Usage Contexts, Noise Cancellation, Features, Quality, and Customer Service. The monthly proportion of sentences that discuss a topic provides the topic trend. Among those six topics, the discussion about the Usage Contexts topic has increased since the beginning of the pandemic, while the other topics do not show a clear trend related to the pandemic. SentiWordNet is selected as the sentiment analysis method to capture the positive and negative sentiment towards the topics. Among the six topics, the Durability Issues and Noise Cancellation topics showed an improved sentiment after the pandemic began, while the sentiment for Usage Contexts, Features, and Quality topics worsened. Future research may be suggested to explain the worsening trend for those topics, especially the Usage Contexts topic that gained significant negativity after the pandemic began. © 2022 IEEE.

Identification of Latent Topics in Patients Surviving COVID-19 in Mexico

With the outbreak of the SARS-CoV-2 o COVID-19 pandemic, multiple studies of risk factors and their influence on patient deaths have been developed. However, little attention is often paid to analyzing patients in risk groups despite the fact that they have been infected and inpatients can survive. In this article, with the dataset available from the Ministery of the health of Mexico, this paper proposes the use of the latent topic extraction algorithm Latent Dirichlet Allocation (LDA) for the study of COVID-19 survival factors in Mexico. The results let us conclude that in the year before strategies for prevention and control of COVID-19, the latent topics support that patients without comorbidities have a low risk of death, compared with the period of 2021, wherein in spite of having some risk factors patients can survive.

Discovering Urban Governance Parameters for Online Learning in Saudi Arabia During COVID-19 Using Topic Modeling of Twitter Data

Smart cities are a relatively recent phenomenon that has rapidly grown in the last decade due to several political, economic, environmental, and technological factors. Data-driven artificial intelligence is becoming so fundamentally ingrained in these developments that smart cities have been called artificially intelligent cities and autonomous cities. The COVID-19 pandemic has increased the physical isolation of people and consequently escalated the pace of human migration to digital and virtual spaces. This paper investigates the use of AI in urban governance as to how AI could help governments learn about urban governance parameters on various subject matters for the governments to develop better governance instruments. To this end, we develop a case study on online learning in Saudi Arabia. We discover ten urban governance parameters using unsupervised machine learning and Twitter data in Arabic. We group these ten governance parameters into four governance macro-parameters namely Strategies and Success Factors, Economic Sustainability, Accountability, and Challenges. The case study shows that the use of data-driven AI can help the government autonomously learn about public feedback and reactions on government matters, the success or failure of government programs, the challenges people are facing in adapting to the government measures, new economic, social, and other opportunities arising out of the situation, and more. The study shows that the use of AI does not have to necessarily replace humans in urban governance, rather governments can use AI, under human supervision, to monitor, learn and improve decision-making processes using continuous feedback from the public and other stakeholders. Challenges are part of life and we believe that the challenges humanity is facing during the COVID-19 pandemic will create new economic, social, and other opportunities nationally and internationally.

A statistical analysis of tweets on covid-19 vaccine hesitancy utilizing opinion mining: an Indian perspective.

The world witnessed the emergence of a deadly virus in December 2019, later named COVID-19. The virus was found to be highly contagious, and so people across the world were highly prone to be affected by the virus. Being a virus-borne disease, developing a vaccine was one of the most promising remedies. Thus, research organizations across the globe started working on developing the vaccine. However, it was later found by many researchers that a large number of people were hesitant to receive the vaccine. This paper aims to study the acceptance and hesitancy levels of people in India and compares them with the acceptance and hesitancy levels of people from the UK, the USA, and the rest of the world by analyzing their tweets on Twitter. For this study, 2,98,452 tweets were fetched from January 2020 to March 2022 from Twitter, and 1,84,720 tweets from 1,22,960 unique users were selected based on their country of origin. Machine learning based Sentiment analysis is then used to evaluate and analyze the tweets. The paper also proposes an NLP-based algorithm to perform opinion mining on Twitter data. The study found the public sentiment of the Indian population to be 63% positive, 28% neutral, and 9% negative. While the worldwide sentiment distribution is 45% positive, 34% neutral, and 21% negative, the USA has 42% positive, 34% neutral, and 23% negative and the UK has 50% positive, 29% neutral, and 21% negative. Also, sentiment analysis for individual vaccines in Indian context resulted in "Covaxin" with the highest positive sentiment at 43% followed by "Covishield" at 36%. The outcome of this work yields an insight into the public perception of the COVID-19 vaccine and thus can be used to formulate policies for existing and future vaccine campaigns. This study becomes more relevant as it is the consolidated opinion of Indian people, which is versatile in nature.

Media Reports on COVID-19 Vaccinations: A Study of Topic Modeling in South Korea.

Early successes in controlling the COVID-19 pandemic have prevented Republic of Korea from implementing a prompt, large-scale vaccine rollout to the public. The influence of traditional media on public opinion remains critical and substantial in Republic of Korea, and there have been heated debates about vaccination in traditional media reports in Korea. Effective and efficient public health communication is integral in managing public health challenges. This study explored media reports on the COVID-19 vaccines during the pandemic in Republic of Korea. 12,399 media news reports from May 2020 to September 2021 were collected. An LDA topic model was applied in order to analyze and compare the topics drawn from each study phase using words from the unstructured text data. Although media reports from before the national vaccination implementation focused on the development and rollout of COVID-19 vaccines, diverse topics were reported without any overlap. After the vaccination rollout, the biggest concern was the side effects of the COVID-19 vaccine. In sum, Republic of Korea's major media outlets reported on diverse topics rather than generating a common discourse about topics related to COVID-19 vaccination.

Analysis of Public Emotion and Disseminator-Audience Framework during Outbreak of Public Crisis Events: A case study of Microblog Account Xiake Island

This study crawled the cross-sectional data of the contents and comments from Microblog Account Xiake Island during the outbreak of coronavirus pneumonia as subjects, to examine the deviation and resonance association among affective fluctuations of the Chinese public, media framework, and audiences' cognitive framework. Using SnowNLP to conduct sentiment analysis of text comments, we found that during the outbreak of coronavirus pneumonia, the public spent most of the time in low-intensity negative affectivity, and the average affective propensity in response to individual microblog fluctuated greatly, and the public was easily caught in an emotional frenzy, which reduces the level of trust in government. Through a comparison of public affectivity and related epidemic data, Xiake Island focuses on reporting emotional facts, whose construction of social reality contains obvious emotional trajectories. Clustering analysis of thematic framework by LDA algorithm reveals that in terms of framework, the framework Xiake Island uses resonates to a large degree with the framework users focus on. In terms of the level of concerns over the framework, Xiake Island deviates to a certain extent from the public. This deviation, together with the strategy of focusing on reporting emotional facts, is a discursive strategy adopted by the new mainstream media to seek the reconstruction of cultural leadership. © 2022 Owner/Author.

Automatic Text Summarization of COVID-19 Scientific Research Topics Using Pre-trained Models from Hugging Face

Automated text summarizing helps the scientific and medical sectors by identifying and extracting relevant information from articles. Automatic text summarization is a way of compressing text documents so that users may find important and useful information in the original text in reduced time. We will first review some new works in the field of summarization that uses deep learning approaches, and then we will explain the application to COVID-19 related research papers. The ease with which a reader can grasp written text is referred to as the readability test. The substance of text determines its readability in natural language processing. We constructed word clouds using the s' most commonly used text. By looking at those three measurements, we can determine the performance measures of ROUGE-1, ROUGE-2, ROUGE-L, ROUGE-L-SUM. Our findings indicated that Distilbart-mnli-12-6 and GPT2-large outperform than others considered. © 2022 IEEE.

COVID-19 fake news analytics from social media using topic modeling and clustering

Nowadays, Health misinformation and myths regarding various types of disease has spread on social media which terrified the public. During COVID-19 pandemic, misinformation and fake news outbreak increased as social media platforms play important role to enable people to view, search, and share the news as well as their point of view globally. Social media users might find difficulties in checking the validity of the news as they could not differentiate which one are the authorized news. Thus, it is too risky if people could easily be swayed by believing the news without validation. Therefore, the goal of this research is to classify the news related to COVID-19 using topic modeling and clustering. Latent Dirichlet Allocation is used for topic modeling of the fake and real news. This study can increase the awareness among social media users to reduce the risk of believing and sharing the misinformation especially during COVID-19 pandemic. © 2022 Elsevier Inc. All rights reserved.

Using Word2Vec-LDA-Word Mover Distance for Comparing the Patterns of Information Seeking and Sharing during the COVID-19 Pandemic

During pandemics such as COVID-19, government announcements were sources to convey accurate and relevant information to the public in times of outbreak. Prior studies attempted to explore the public awareness and behavioral changes from various research disciplines in response to the COVID-19 pandemic. Literature has pointed out that the appropriate use of information sources significantly relates to public attitudes in battling the pandemic. Social media has been the widely used medium to express public interests in current events. Literature shows that social media use during a crisis effectively coordinates relevant information from different sources and promotes situational awareness. Therefore, it is crucial to investigate scalable approaches to promptly gather insights into the public's interests and how governments responded to the interests relevant to the COVID-19 pandemic. However, there is little empirical research found that tackles these needs. Therefore, we aim to close the research gap by examining the feasible approaches for (1) identifying if public information-seeking has similar patterns as information-sharing on social media during the COVID-19 pandemic, and (2) comparing the patterns with the government announcements to confirm if the announcements show aligned response to the public information-seeking and sharing during the COVID-19 pandemic. We applied text processing, LDA topic modeling, and Word Mover Distance techniques to realize our aim through a Malaysian case study. Our research work contributes to the application of the LDA-Word2Vec-Word Mover Distance architecture and algorithms that can be used for future investigation and comparison of information seeking and sharing patterns in different research subjects. © 2022 IEEE.

KEAHT: A Knowledge-Enriched Attention-Based Hybrid Transformer Model for Social Sentiment Analysis.

Social media materialized as an influential platform that allows people to share their views on global and local issues. Sentiment analysis can handle these massive amounts of unstructured reviews and convert them into meaningful opinions. Undoubtedly, COVID-19 originated as the enormous challenge across the world that physically and financially bruted humankind. Meanwhile, farmers' protests shook up the world against three pieces of legislation passed by the Indian government. Hence, an artificial intelligence-based sentiment model is needed for suggesting the right direction toward outbreaks. Although Deep Neural Network (DNN) gained popularity in sentiment analysis applications, these still have a limitation of sequential training, high-dimension feature space, and equal feature importance distribution. In addition, inaccurate polarity scoring and utility-based topic modeling are other challenging aspects of sentiment analysis. It motivates us to propose a Knowledge-Enriched Attention-based Hybrid Transformer (KEAHT) model by enriching the explicit knowledge of Latent Dirichlet Allocation (LDA) topic modeling and lexicalized domain ontology. A pre-trained Bidirectional Encoder Representation from Transformer (BERT) is employed to train within a minimum training corpus. It provides the facility of attention mechanism and can solve complex text problems accurately. A comparative study with existing baselines and recent hybrid models affirms the credibility of the proposed KEAHT in the field of Natural Language Processing (NLP). This model emphasizes artificial intelligence's role in handling the situation of the global pandemic and democratic dispute in a country. Furthermore, two benchmark datasets, namely "COVID-19-Vaccine-Labelled-Tweets" and "Indian-Farmer-Protest-Labelled-Tweets", are also constructed to accommodate future researchers for outlining the essential facts associated with the outbreaks.

Semantic Analysis and Topic Modelling of Web-Scrapped COVID-19 Tweet Corpora through Data Mining Methodologies.

The evolution of the coronavirus (COVID-19) disease took a toll on the social, healthcare, economic, and psychological prosperity of human beings. In the past couple of months, many organizations, individuals, and governments have adopted Twitter to convey their sentiments on COVID-19, the lockdown, the pandemic, and hashtags. This paper aims to analyze the psychological reactions and discourse of Twitter users related to COVID-19. In this experiment, Latent Dirichlet Allocation (LDA) has been used for topic modeling. In addition, a Bidirectional Long Short-Term Memory (BiLSTM) model and various classification techniques such as random forest, support vector machine, logistic regression, naive Bayes, decision tree, logistic regression with stochastic gradient descent optimizer, and majority voting classifier have been adapted for analyzing the polarity of sentiment. The effectiveness of the aforesaid approaches along with LDA modeling has been tested, validated, and compared with several benchmark datasets and on a newly generated dataset for analysis. To achieve better results, a dual dataset approach has been incorporated to determine the frequency of positive and negative tweets and word clouds, which helps to identify the most effective model for analyzing the corpora. The experimental result shows that the BiLSTM approach outperforms the other approaches with an accuracy of 96.7%.

Musawah: A Data-Driven AI Approach and Tool to Co-Create Healthcare Services with a Case Study on Cancer Disease in Saudi Arabia

The sustainability of human existence is in dire danger and this threat applies to our environment, societies, and economies. Smartization of cities and societies has the potential to unite individuals and nations towards sustainability as it requires engaging with our environments, analyzing them, and making sustainable decisions regulated by triple bottom line (TBL). Poor healthcare systems affect individuals, societies, the planet, and economies. This paper proposes a data-driven artificial intelligence (AI) based approach called Musawah to automatically discover healthcare services that can be developed or co-created by various stakeholders using social media analysis. The case study focuses on cancer disease in Saudi Arabia using Twitter data in the Arabic language. Specifically, we discover 17 services using machine learning from Twitter data using the Latent Dirichlet Allocation algorithm (LDA) and group them into five macro-services, namely, Prevention, Treatment, Psychological Support, Socioeconomic Sustainability, and Information Availability. Subsequently, we show the possibility of finding additional services by employing a topical search over the dataset and have discovered 42 additional services. We developed a software tool from scratch for this work that implements a complete machine learning pipeline using a dataset containing over 1.35 million tweets we curated during September–November 2021. Open service and value healthcare systems based on freely available information can revolutionize healthcare in manners similar to the open-source revolution by using information made available by the public, the government, third and fourth sectors, or others, allowing new forms of preventions, cures, treatments, and support structures.

Trend Analysis of Pet Plants Before and After COVID-19 Outbreak Using Topic Modeling: Focusing on Big Data of News Articles from 2018 to 2021

Background and objective: The ongoing COVID-19 pandemic restricted daily life, forcing people to spend time indoors. With the growing interest in mental health issues and residential environments, ‘pet plants’ have been receiving attention during the unprecedented social distancing measures. This study aims to analyze the change in trends of pet plants before and during the COVID-19 pandemic and provide basic data for studies related to pet plants and directions of future development. Methods: A total of 2,016 news articles using the keyword ‘pet plants’ were collected on Naver News from January 1, 2018 to August 15, 2019 (609 articles) and January 1, 2020 to August 15, 2021 (1,407 articles). The texts were tokenized into words using KoNLPy package, ultimately coming up with 63,597 words. The analyses included frequency of keywords and topic modeling based on Latent Dirichlet Allocation (LDA) to identify the inherent meanings of related words and each topic. Results: Topic modeling generated three topics in each period (before and during the COVID-19), and the results showed that pet plants in daily life have become the object of ‘emotional support’ and ‘healing’ during social distancing. In particular, pet plants, which had been distributed as a solution to prevent solitary deaths and depression among seniors living alone, are now expanded to help resolve the social isolation of the general public suffering from COVID-19. The new term ‘plant butler’ became a new trend, and there was a change in the trend in which people shared their hobbies and information about pet plants and communicated with others in online. Conclusion: Based on these findings, the trend data of pet plants before and after the outbreak of COVID-19 can provide the basis for activating research on pet plants and setting the direction for development of related industries considering the continuous popularity and trend of indoor gardening and green hobby. © 2021 by the Society for People, Plants, and Environment.

Using Word2Vec-LDA-Cosine Similarity for Discovering News Dissemination Pattern to Support Government–Citizen Engagement

Evidence of ineffective government–citizen engagement was observed when the Malaysian government decided to make face masks mandatory in public spaces. It is especially critical during a COVID-19 pandemic, where public compliance depends on the speed and clarity at which regulations are announced. Hundreds of arrested cases due to violation were met with confusion and demanded greater clarification. This evidence signifies the need to identify if government-disseminated information is communicated effectively to the citizens through news coverage. Despite this need, current literature has limitations in effectively analysing huge numbers of articles as they mainly employ manual intervention for data news content analysis. Furthermore, there has been no usage of systematic text analytics approaches in government–citizen engagement studies through newspapers. As such, we researched and implemented a modelling framework for discovering how news coverage pattern aligns with government-disseminated information through a case study of COVID-19 in Malaysia during the pandemic. A Word2Vec-LDA-cosine similarity technique was employed in our framework to determine topic similarities as the indication of alignment between news content and government-disseminated information. Our results show that this framework succeeds in capturing the semantics of the corpus to describe news coverage at the same time identified the challenges in general topic comparison tasks. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Spatial evolution patterns of public panic on Chinese social networks amidst the COVID-19 pandemic.

Novel coronavirus pneumonia has had a significant impact on people's lives and psychological health. We developed a stage model to analyse the spatial and temporal distribution of public panic during the two waves of the coronavirus disease 2019 (COVID-19) pandemic. We used tweets with geographic location data from the popular hashtag 'Lockdown Diary' recorded from 23 January to April 8, 2020, and 'Nanjing Outbreak' recorded from 21 July to 1 September 2021 on Weibo. Combining the lexicon-based sentiment analysis and the grounded theory approach, this panic model could explain people's panic and behavioural responses in different areas at different stages of the pandemic. Next, we used the latent Dirichlet allocation topic model to reconfirm the panic model. The results showed that public sentiments fluctuated strongly in the early stages; in this case, panic and prayers were the dominant sentiments. In terms of spatial distribution, public panic showed hierarchical and neighbourhood diffusion, with highly assertive expressions of sentiment at the outbreak sites, economically developed areas, and areas surrounding the outbreak. Most importantly, we considered that public panic was affected by the 17 specific topics extracted based on the perceived and actual distance of the pandemic, thus stimulating the process of panic from minimal, acute, and mild panic to perceived rationality. Consequently, the public's behavioural responses shifted from delayed, negative, and positive, to rational behavioural responses. This study presents a novel approach to explore public panic from both a time and space perspective and provides some suggestions in response to future pandemics.

Internet Public Opinion Evolution in the COVID-19 Event and Coping Strategies.

OBJECTIVES: In this study, we carried out a text analysis on the information disseminated and discussed among netizens on the Baidu Post Bar (the world's largest Chinese forum) during the coronavirus disease 2019 (COVID-19) epidemic, to create a policy basis for health administrative departments. METHODS: We used Python tools to search for the relevant data on the Baidu Post Bar. Next, a text analysis was performed on the posts' contents using a combination of latent Dirichlet allocation (LDA), sentiment analysis, and correlation analysis. RESULTS: According to the LDA analysis, the public was highly interested in topics such as COVID-19 prevention, infection symptoms, infection and coping measures, sources of transmission and treatments, community management, and work resumption. The majority of the public had negative emotional values, yet a portion of the public held positive emotional values. We also performed a correlation analysis of the influencing factors was established. CONCLUSIONS: Netizens' degree of concern shown in their posts was greatly associated with the spread of COVID-19. With the rise, diffusion, outbreak, and mitigation of COVID-19 in China, netizens have successively created a large number of posts, and the topics of discussion varied over time. Therefore, the media and the government have the responsibility to distribute positive information, to correctly guide the public's emotions to bring some sort of reassurance to the public.