Search

Despite evidence indicating the significant influence of online reviews on purchase decisions, even after considering a product's average rating (Vana and Lambrecht 2021), the underlying factors behind this effect and the broader impact of reviews on consumer decision making remain uncertain. This study uses clickstream data from a major online retailer to explore how negative reviews affect consumer search and purchase decisions. Leveraging exogenous variation created by the display of online reviews sorted by recency, the authors find that negative reviews significantly reduce a product's purchase probability because they (1) contrast with the often-high average product rating, (2) decrease the probability that consumers continue browsing for information about the focal product, (3) increase the probability of visiting the page of substitute products, and (4) increase the probability of viewing reviews about substitute products. Importantly, these effects apply to utilitarian products but not hedonic products and when reviews pertain to product functionality or customer service but not to taste-related factors. The authors estimate a product's vulnerability to negative reviews along two dimensions—purchase and search probability for substitutes—and display these effects on a two-dimensional map. [ABSTRACT FROM AUTHOR]

The authors examine the effects of a firm's and its competitors' online reviews on its demand within the hotel industry. The authors leverage a unique data set of actual bookings from properties of a major hotel chain in six different markets in the United States, supplemented with online reviews garnered from a popular social media platform. The findings indicate that not only a hotel's own reviews but also its competitors' reviews have a significant impact on the hotel's booking performance. The impact of review sentiment is amplified if the focal hotel also charges higher prices or when the volume of reviews is high. The authors establish heterogeneous effects across consumer segments (business vs. leisure travelers) and by the type of review content (objective vs. subjective attributes to assess quality). Specifically, both a hotel's own reviews and its competitors' reviews have a larger impact on bookings for business travelers compared with leisure travelers, and for reviews that mainly discuss subjective attributes, for which consumers need to rely on the experiences of others to assess the quality of a hotel prior to their stay. The study provides a set of comprehensive insights on the impact of both own and competitors' online reviews on a focal hotel's bookings. [ABSTRACT FROM AUTHOR]

ObjectiveExisting research on social determinants of health (SDoH) predominantly focuses on physician notes and structured data within electronic medical records. This study posits that social work notes are an untapped, potentially rich source for SDoH information. We hypothesize that clinical notes recorded by social workers, whose role is to ameliorate social and economic factors, might provide a complementary information source of data on SDoH compared to physician notes, which primarily concentrate on medical diagnoses and treatments. We aimed to use word frequency analysis and topic modeling to identify prevalent terms and robust topics of discussion within a large cohort of social work notes including both outpatient and in-patient consultations.Materials and methodsWe retrieved a diverse, deidentified corpus of 0.95 million clinical social work notes from 181 644 patients at the University of California, San Francisco. We conducted word frequency analysis related to ICD-10 chapters to identify prevalent terms within the notes. We then applied Latent Dirichlet Allocation (LDA) topic modeling analysis to characterize this corpus and identify potential topics of discussion, which was further stratified by note types and disease groups.ResultsWord frequency analysis primarily identified medical-related terms associated with specific ICD10 chapters, though it also detected some subtle SDoH terms. In contrast, the LDA topic modeling analysis extracted 11 topics explicitly related to social determinants of health risk factors, such as financial status, abuse history, social support, risk of death, and mental health. The topic modeling approach effectively demonstrated variations between different types of social work notes and across patients with different types of diseases or conditions.DiscussionOur findings highlight LDA topic modeling's effectiveness in extracting SDoH-related themes and capturing variations in social work notes, demonstrating its potential for informing targeted interventions for at-risk populations.ConclusionSocial work notes offer a wealth of unique and valuable information on an individual's SDoH. These notes present consistent and meaningful topics of discussion that can be effectively analyzed and utilized to improve patient care and inform targeted interventions for at-risk populations.

IntroductionThe rise in global temperatures due to climate change has escalated the frequency and intensity of wildfires worldwide. Beyond their direct impact on physical health, these wildfires can significantly impact mental health. Conventional mental health studies predominantly rely on surveys, often constrained by limited sample sizes, high costs, and time constraints. As a result, there is an increasing interest in accessing social media data to study the effects of wildfires on mental health.MethodsIn this study, we focused on Twitter users affected by the California Tubbs Fire in 2017 to extract data signals related to emotional well-being and mental health. Our analysis aimed to investigate tweets posted during the Tubbs Fire disaster to gain deeper insights into their impact on individuals. Data were collected from October 8 to October 31, 2017, encompassing the peak activity period. Various analytical methods were employed to explore word usage, sentiment, temporal patterns of word occurrence, and emerging topics associated with the unfolding crisis.ResultsThe findings show increased user engagement on wildfire-related Tweets, particularly during nighttime and early morning, especially at the onset of wildfire incidents. Subsequent exploration of emotional categories using Linguistic Inquiry and Word Count (LIWC) revealed a substantial presence of negative emotions at 43.0%, juxtaposed with simultaneous positivity in 23.1% of tweets. This dual emotional expression suggests a nuanced and complex landscape, unveiling concerns and community support within conversations. Stress concerns were notably expressed in 36.3% of the tweets. The main discussion topics were air quality, emotional exhaustion, and criticism of the president's response to the wildfire emergency.DiscussionSocial media data, particularly the data collected from Twitter during wildfires, provides an opportunity to evaluate the psychological impact on affected communities immediately. This data can be used by public health authorities to launch targeted media campaigns in areas and hours where users are more active. Such campaigns can raise awareness about mental health during disasters and connect individuals with relevant resources. The effectiveness of these campaigns can be enhanced by tailoring outreach efforts based on prevalent issues highlighted by users. This ensures that individuals receive prompt support and mitigates the psychological impacts of wildfire disasters.

Understanding the discrepancy between passenger expectations and airline offerings is critical for improving service quality and customer satisfaction. This study expands the current knowledge regarding airline service quality dimensions by investigating actual passenger experiences. Analyzing 1,389,689 online reviews from TripAdvisor.com, structural topic modeling identified 29 airline service quality factors. Drawing upon AIRQUAL and SERVQUAL as theoretical frameworks, five unique dimensions emerged: flight attendance service, flight schedule, in-flight entertainment, in-flight meal, and baggage. Moreover, regression analysis highlights the significance of flight attendant services and hidden critical factors, such as attention to child passengers and special requests, in enhancing customer satisfaction. This study provides theoretical implications by uncovering airline service quality factors recognized by passengers and suggesting five potential service quality dimensions suitable for the airline context. Practically, the current study guides airline managers in understanding a passenger viewpoint and avoiding the service gap by combining machine learning, econometric approaches, and existing theories. [ABSTRACT FROM AUTHOR]

Lung cancer (LC) remains one of the leading causes of cancer‐related mortality worldwide. With recent technological advances, artificial intelligence (AI) has begun to play a crucial role in improving diagnostic and treatment methods. It is crucial to understand how AI has integrated into LC research and to identify the main areas of focus. The aim of the study was to provide an updated insight into the role of AI in LC research, analyzing evolving topics, geographical distribution, and contributions to journals. The study explores research trends in AI applied to LC through a novel approach combining latent Dirichlet allocation (LDA) topic modeling with the HJ‐Biplot statistical technique. A growing interest in AI applications in LC oncology was observed, reflected in a significant increase in publications, especially after 2017, coinciding with the availability of computing resources. Frontiers in Oncology leads in publishing AI‐related LC research, reflecting rigorous investigation in the field. Geographically, China and the United States lead in contributions, attributed to significant investment in R&D and corporate sector involvement. LDA analysis highlights key research areas such as pulmonary nodule detection, patient prognosis prediction, and clinical decision support systems, demonstrating the impact of AI in improving LC outcomes. DL and AI emerge as prominent trends, focusing on radiomics and feature selection, promising better decision‐making in LC care. The increase in AI‐driven research covers various topics, including data analysis methodologies, tumor characterization, and predictive methods, indicating a concerted effort to advance LC research. HJ‐Biplot visualization reveals thematic clustering, illustrating temporal and geographical associations and highlighting the influence of high‐impact journals and countries with advanced research capabilities. This multivariate approach offers insights into global collaboration dynamics and specialization, emphasizing the evolving role of AI in LC research and diagnosis. [ABSTRACT FROM AUTHOR]

This study analyzes the status, trends, and future directions in wood science research using text-mining techniques. We applied these techniques to a textual dataset constructed from metadata of six major wood science journals, covering the period from 2002 to 2024. The research explores publication trends, international collaborations, keywords, and research networks, and it employs topic modeling using the Latent Dirichlet Allocation model. The descriptive analysis reveals a consistent increase in publication volume throughout the study period, unaffected by the COVID-19 pandemic. In contrast, international collaboration declined after 2020, likely due to the pandemic. In addition, a network analysis identified key research areas, including surface treatments, structural composites, and high-performance wood products, with lignin, mechanical properties, and moisture content emerging as central keywords. Topic modeling reveals a growing interest in wood modification technologies and an increased focus on studying wood as a sustainable material. The study confirms a shift of the field towards sustainable innovations while also highlighting the enduring relevance of traditional research areas. Future research should adapt to these evolving trends and address emerging challenges to maximize the potential of wood for carbon neutrality and sustainable development. This analysis provides a concise overview of current research trends and future directions in wood science. [ABSTRACT FROM AUTHOR]

Introduction: The integration of artificial intelligence (AI) in pathogenic microbiology has accelerated research and innovation. This study aims to explore the evolution and trends of AI applications in this domain, providing insights into how AI is transforming research and practice in pathogenic microbiology. Methods: We employed bibliometric analysis and topic modeling to examine 27,420 publications from the Web of Science Core Collection, covering the period from 2010 to 2024. These methods enabled us to identify key trends, research areas, and the geographical distribution of research efforts. Results: Since 2016, there has been an exponential increase in AI-related publications, with significant contributions from China and the USA. Our analysis identified eight major AI application areas: pathogen detection, antibiotic resistance prediction, transmission modeling, genomic analysis, therapeutic optimization, ecological profiling, vaccine development, and data management systems. Notably, we found significant lexical overlaps between these areas, especially between drug resistance and vaccine development, suggesting an interconnected research landscape. Discussion: AI is increasingly moving from laboratory research to clinical applications, enhancing hospital operations and public health strategies. It plays a vital role in optimizing pathogen detection, improving diagnostic speed, treatment efficacy, and disease control, particularly through advancements in rapid antibiotic susceptibility testing and COVID-19 vaccine development. This study highlights the current status, progress, and challenges of AI in pathogenic microbiology, guiding future research directions, resource allocation, and policy-making. [ABSTRACT FROM AUTHOR]

Hans Christian Andersen's fairy tales have garnered the greatest popular and scholarly attention despite the interdependence of works across the broad range of his artistic production. We read Andersen's fairy tales in concert with his travel writing to highlight the intertextual aspects that cross these seemingly distinct genres. We leverage recent advances in large language models (LLM) and network theory to generate representations that facilitate user exploration of these intertextual interdependencies across genres and across time. In the first part of our study, we use BERTopic and an LLM model fine‐tuned for nineteenth‐century Danish literary language to present independent and combined topic models of the two corpuses. This approach supports multi‐scalar analysis of intertextual elements within and across these corpuses, thereby implementing a method for macroscopic reading. In the second part of the study, we develop a series of networked representations of the dependencies between fairy tales, where these dependencies are generated on the basis of the shared intertextual topic space of the fairy tales and the travel writing. [ABSTRACT FROM AUTHOR]

Chinese migrant workers have contributed greatly to the industrialization and urbanization of China and have thus long been a popular topic in news reports. Most previous studies on the news about this group have been qualitative or quantitative analyses based on small-scale datasets, with few based on large-scale corpus data. This study combined topic modeling and critical discourse analysis (CDA) to analyze 186,180 reports of migrant workers published by 906 Chinese newspapers between 2001 and 2021. We identified seventeen topics in the news that fell into five categories: "work and labor," "support and care," "rights and interests," "homecoming," and "social emergencies." Overall, news reports on the group have become increasingly diverse, with various diachronic changes in the proportions of topic categories and specific topics. In addition, we found three major types of representation of migrant workers: as a disadvantaged group, as outsiders, and as a group vulnerable to risks. These negative representations might be closely related to government policies, ideology, media stance, and social reality (the broader societal context in which these individuals are situated). [ABSTRACT FROM AUTHOR]

Systematic reviews are a time-consuming yet effective approach to understanding research trends. While researchers have investigated how to speed up the process of screening studies for potential inclusion, few have focused on to what extent we can use algorithms to extract data instead of human coders. In this study, we explore to what extent analyses and algorithms can produce results similar to human data extraction during a scoping review—a type of systematic review aimed at understanding the nature of the field rather than the efficacy of an intervention—in the context of a never before analyzed sample of studies that were intended for a scoping review. Specifically, we tested five approaches: bibliometric analysis with VOSviewer, latent Dirichlet allocation (LDA) with bag of words, k-means clustering with TF-IDF, Sentence-BERT, or SPECTER, hierarchical clustering with Sentence-BERT, and BERTopic. Our results showed that topic modeling approaches (LDA/BERTopic) and k-means clustering identified specific, but often narrow research areas, leaving a substantial portion of the sample unclassified or in unclear topics. Meanwhile, bibliometric analysis and hierarchical clustering with SBERT were more informative for our purposes, identifying key author networks and categorizing studies into distinct themes as well as reflecting the relationships between themes, respectively. Overall, we highlight the capabilities and limitations of each method and discuss how these techniques can complement traditional human data extraction methods. We conclude that the analyses tested here likely cannot fully replace human data extraction in scoping reviews but serve as valuable supplements. [ABSTRACT FROM AUTHOR]

Increasingly, influencers are employed to market not only products but also ideas and beliefs. The far right has recognized the strategic potential of influencer communication to tap into new target groups and mobilize supporters. This paper provides insights into the little-explored field of far-right influencers. We conceptualize them as individual actors characterized by far-right ideology, positioned as political influencers, actively advocating for their ideological aims. Employing a multi-layered computational approach to explore communication practices and networking structures of 243 German-speaking far-right influencers on Telegram, we derive a typology and observe the emergence of a functionally differentiated influencer collective. In this collective, each community has specific functions and characteristics that emphasize different ideological aspects, mobilization modes, and influencer practices. Despite the decentralized organization, we find high efficiency in information dissemination. The results corroborate the assumed potential of far-right influencers as disseminators of ideological content who can be particularly persuasive through their role as parasocial opinion leaders. [ABSTRACT FROM AUTHOR]

This study uses topic modeling to analyze research on burnout among firefighters, police, and first responders. The corpus contains 148 abstracts and 29,517 words. Latent Dirichlet Allocation identified 8 key topics, including mental health factors like PTSD, substance abuse, and suicide, as well as organizational factors like job demands and social support. The study found limited focus on prevention and cultural nuances. It proposes developing comprehensive models incorporating lesser-known factors and subtle cultural contexts to better understand and reduce burnout in these critical professions. [ABSTRACT FROM AUTHOR]

Objective: To analyze Chinese medicine (CM) prescriptions for gastroesophageal reflux disease (GERD), we model topics on GERD-related classical CM literature, providing insights into the potential treatment. Methods: Clinical guidelines were used to identify symptom terms for GERD, and CM literature from the database "Imedbooks" was retrieved for related prescriptions and their corresponding sources, indications, and other information. BERTopic was applied to identify the main topics and visualize the data. Results: A total of 36,207 entries are queried and 1,938 valid entries were acquired after manually filtering. Eight topics were identified by BERTopic, including digestion function abate, stomach flu, respiratory-related symptoms, gastric dysfunction, regurgitation and gastrointestinal dysfunction in pediatric patients, vomiting, stroke and alcohol accumulation are associated with the risk of GERD, vomiting and its causes, regurgitation, epigastric pain, and symptoms of heartburn. Conclusions: Topic modeling provides an unbiased analysis of classical CM literature on GERD in a time-efficient and scale-efficient manner. Based on this analysis, we present a range of treatment options for relieving symptoms, including herbal remedies and non-pharmacological interventions such as acupuncture and dietary therapy. [ABSTRACT FROM AUTHOR]

The proliferation of the internet has provided many tools and platforms for consumers to share and create content concerning their experiences with different products and services. Customer reviews are considered essential both in strategy development and consumer attraction. By analyzing customer reviews, companies can decide on product enhancement and development, service recovery and improvements, pricing, and customer recruitment and retention. Although different methodologies have been used to understand the impact of customer reviews in the last decade, a new realm of studies began to benefit from text analytics and machine learning algorithms. Within these methodologies, topic modeling algorithms emerge as the most popular analytical tool. However, despite the subsistence of different topic modeling algorithms, their exposure in marketing research has been limited. This study aims to provide information on different algorithms' semantic potential in analyzing customer reviews. Using a comprehensive dataset extracted from the Best Buy platform, the contributions of various algorithms are compared based on data preprocessing, algorithm implementation, and their semantic ability in marketing research. Based on different stages of implementation and their respective complexity, the results indicate efficacy in using BERTopic in analyzing customer reviews. [ABSTRACT FROM AUTHOR]

Digital and sharing platforms are transforming the business landscape and significantly impacting the global economy. This study provides a systematic review of the sharing economy (SE) in marketing and presents a future research agenda to advance the discipline. A comprehensive approach is utilized, incorporating bibliometric analysis, performance analysis, science mapping techniques, and topic modeling to analyze sharing economy research within marketing. The dataset comprises 325 articles on the sharing economy, from which eight key topics are identified: (1) analyzing customer ratings and sentiments, (2) understanding and evaluating customer experience, (3) customer orientation towards peer-to-peer accommodation services, (4) mapping customer relationships in the sharing economy, (5) analysis of user loyalty, (6) development of marketing strategies to enhance platforms, (7) identification of platform openness, and (8) actors within the digital platform ecosystem. By identifying research gaps, a future research model is proposed to comprehend the roles and influence of customers, peer service providers, and platform providers in the sharing economy. The findings of this study have practical implications for marketing practices within the SE. [ABSTRACT FROM AUTHOR]

How have urban planners addressed racial equity and justice in their scholarship? Using topic modeling and qualitative text coding, this study analyzed articles from twenty top urban planning journals from 1980 to 2019. The findings indicate that only 1.8 percent of planning articles are related to racial equity and justice throughout the four decades, with a slight uptick in attention paid to these topics in recent years. The dearth of planning scholarship on these topics in high-impact journals suggests that planning, as an academic discipline, has not prioritized and centered research and teaching about racial equity and justice. [ABSTRACT FROM AUTHOR]

Collusive practices continue to be a significant threat to competition and consumer welfare. It should be of utmost importance for academic research to provide the theoretical and empirical foundations to antitrust authorities and enable them to develop proper tools to encounter new collusive practices. Utilizing topical natural language machine learning techniques allows me to analyze the evolution of economic research on collusion over the past two decades in a novel way. It enables me to review some 800 publications systematically. I extract the underlying topics from the papers and conduct a large set of uni‐ and multivariate time series and regression analyses on their individual prevalences. I detect a notable tendency towards monocultures in topics and an endogenous constriction of the topic variety. In contrast, the overall contents and issues addressed by these papers have grown remarkably. This caused a decoupling: Nowadays, more datasets and cartel cases are studied but with a smaller research scope. [ABSTRACT FROM AUTHOR]

The research on immigrant integration is a dynamic interdisciplinary domain with rich and diverse literature. Considering the numerous studies in this domain, it is valuable to provide a comprehensive mapping to understand the research landscape resource and facilitate collaborations. This article combines Structural Topic modeling with bibliometric analysis to identify key research topics on immigrant integration. Applying these methods to 70890 abstracts published between 1960 and 2022, we identified 30 key research topics. We also tracked their prevalence over time, correlations and distributions within institutions and countries, funding effects, and countries' collaborations. The results indicate that the most discussed topics are integration theory, economic integration, education, residential segregation, integration policy, language, religious diversity, cultural participation, identity & belonging, racism & discrimination, political participation, health & welfare, research methods, demographic issues, gender & violence. Moreover, some of these topics were highly prominent in earlier periods and nearly non-existent in later years, while others emerged only recently. Specific topics maintained consistent significance over time. Analyses of correlations and trends reveal clusters of topics and diverse distributions across countries and research institutions. The implications of these results can benefit researchers, helping them better understand the current state of research and design future research projects. The finding could also help stakeholders in migration and integration governance and funding agencies to guide policies regarding the integration of immigrants. [ABSTRACT FROM AUTHOR]

Dark Web has turned into a platform for a variety of criminal activities, including weapon trafficking, pornography, fake documents, drug trafficking, and, most notably terrorism as detailed in this study. This article uses an LDA-based topic modeling approach to identify the topics addressed in discussions on the Dark Web. The main purpose is to present an overview of jihadists' communication in cyberspace for the detection of unusual behavior or terrorism-related purposes. According to the findings, conversations in the context of recruitment and propaganda predominated at the forum. There was no direct evidence of terrorist collaboration at the conclusion of the investigation. This does not, however, imply that these sites are risk-free. Propaganda and recruitment tools feed the terrorist activities. [ABSTRACT FROM AUTHOR]

This study aims to reveal the competence areas and skill sets needed in the business world in the field of cyber security (CS). For this purpose, descriptive analysis, topic modeling analysis, and semantic content analysis were conducted on 9407 CS job advertisements obtained from Indeed.com. The results of the study revealed a total of 10 job titles and 23 skill sets expected by the business world in CS job postings. The first three titles in terms of volume were "Engineer", "Analyst" and "Specialist", while the first three skill sets were determined as "Business/Customer services", "System engineering", and "Bachelor degree". In addition, maps showing the relationships between titles and skills were created with a title-skill set matrix. The results of our study can be expected to contribute to candidates and professionals in the field of cyber security, IT organizations, and educational institutions in the cyber security business world by seeing, evaluating, developing, and expanding the current knowledge, skills, and competencies needed in the field. [ABSTRACT FROM AUTHOR]

Technology has substantially transformed the way legal services operate in many different countries. With a large and complex collection of digitized legal documents, the judiciary system worldwide presents a promising scenario for the development of intelligent tools. In this work, we tackle the challenging task of organizing and summarizing the constantly growing collection of legal documents, uncovering hidden topics, or themes that later can support tasks such as legal case retrieval and legal judgment prediction. Our approach to this problem relies on topic discovery techniques combined with a variety of preprocessing techniques and learning-based vector representations of words, such as Doc2Vec and BERT-like models. The proposed method was validated using four different datasets composed of short and long legal documents in Brazilian Portuguese, from legal decisions to chapters in legal books. Analysis conducted by a team of legal specialists revealed the effectiveness of the proposed approach to uncover unique and relevant topics from large collections of legal documents, serving many purposes, such as giving support to legal case retrieval tools and also providing the team of legal specialists with a tool that can accelerate their work of labeling/tagging legal documents. [ABSTRACT FROM AUTHOR]

Autism spectrum disorder (ASD) is characterized by highly heterogenous abnormalities in functional brain connectivity affecting social behavior. There is a significant progress in understanding the molecular and genetic basis of ASD in the last decade using multi-omics approach. Mining this large volume of biomedical literature for insights requires considerable amount of manual intervention for curation. Machine learning and artificial intelligence fields are advancing toward simplifying data mining from unstructured text data. Here, we demonstrate our literature mining pipeline to accelerate data to insights. Using topic modeling and generative AI techniques, we present a pipeline that can classify scientific literature into thematic clusters and can help in a wide array of applications such as knowledgebase creation, conversational virtual assistant, and summarization. Employing our pipeline, we explored the ASD literature, specifically around multi-omics studies to understand the molecular interplay underlying autism brain. [ABSTRACT FROM AUTHOR]

The present study provides an overview of syntactic complexity (SC) in second language (L2) writing research, employing the Latent Dirichlet Allocation (LDA) topic modeling approach. Analyzing 470 abstracts sourced from the WoS Core Collection (2005–2023) and Scopus (1978–2023), this investigation explores the broad spectrum of research themes in SC in L2 writing. By employing perplexity and coherence tests, the study identifies the optimal number of topics in the model as five: writing quality, proficiency, genres, SC development, and task-based conditions. The finding reveals that: (1) In cross-sectional studies, the interplay among writing quality, proficiency levels, and genres often involves studying one variable as reliant on the others. Additionally, there is a growing inclination toward regarding phrasal structures as more dependable complexity indicators. (2) Longitudinal investigations predominantly emphasize the evolution of SC, yet there’s a scarcity of exploration in continuation writing tasks. (3) Regarding measurement metrics, a transition from broader SC evaluations to more intricate assessments is evident, though several studies lack thorough examinations, particularly in continuation tasks. [ABSTRACT FROM AUTHOR]

This study examines higher education through data-mining methodologies, aiming to uncover key themes and sentiments in global discourse. Utilizing sentiment analysis and topic modeling, the research analyzes 157,943 tweets from 84,423 unique users over a five-month period (January to May 2023). This period was selected, coinciding with the rise of artificial intelligence (AI) tools, particularly ChatGPT. The study investigates the discussions, emotional tones, and dominant topics shaping the global narrative of higher education within X (Twitter) data. Key findings include the geographical distribution of tweets and the most frequent positive and negative perceptions. It also addresses critical issues such as affordability, accessibility, and funding in higher education. Furthermore, the data shows public reactions to AI in higher education are initially negative, while higher education tweets are primarily characterized by positivity and optimism. The higher education tweets are mainly posted on the weekend, with decreased activity during weekdays. This research provides insights into the evolving higher education landscape amid rapid technological advancements. [ABSTRACT FROM AUTHOR]

AbstractBackgroundMethodsResultsConclusionNicotine gum products from brands like Lucy and Rogue are relatively new arrivals to the tobacco marketplace. While studies of correlates of nicotine gum use are in their nascent stage, data from social media can be used to stay abreast of user experiences with novel tobacco products. This study leveraged machine learning to identify topics of Twitter posts about nicotine gum from the year 2022.Twitter data was collected using the Twitter Application Programming Interface (API), with search terms “nicotine” AND “gum” OR “nic” AND “gum”. 16,940 tweets from 10,353 unique users were included in the analysis. Topic modeling with Top2Vec was used to identify topics and a string search of popular brands and flavors was also conducted.Eight distinct topics were identified. Smoking and vaping cessation was the most common topic, followed by promotion, pricing, and marketing; appeal; product comparisons; perceived benefits; distrust in institutions; health concerns; and COVID-19 misinformation. The most mentioned brand was Nicorette. A tenth of the tweets contained a reference to a flavor.The goal of identifying topics in nicotine gum-related conversations is to better understand the public’s perceptions and experiences with nicotine gum. These findings may be used to inform survey-based research, policy targets, and health communication campaigns. [ABSTRACT FROM AUTHOR]

In this paper, we propose TAWE (title assignment with word embedding), a new method to automatically assign titles to topics inferred from sets of documents. This method combines the results obtained from the topic modeling performed with, e.g., latent Dirichlet allocation (LDA) or other suitable methods and the word embedding representation of words in a vector space. This representation preserves the meaning of the words while allowing to find the most suitable word that represents the topic. The procedure is twofold: first, a cleaned text is used to build the LDA model to infer a desirable number of latent topics; second, a reasonable number of words and their weights are extracted from each topic and represented in n-dimensional space using word embedding. Based on the selected weighted words, a centroid is computed, and the closest word is chosen as the title of the topic. To test the method, we used a collection of tweets about climate change downloaded from some of the main newspapers accounts on Twitter. Results showed that TAWE is a suitable method for automatically assigning a topic title. [ABSTRACT FROM AUTHOR]

The current research adopted a network‐analytic approach to summarize the changes within the field of personality psychology over a 30‐year span from 1990 to 2019. Bibliographic data from 25,238 articles were used to construct three separate co‐authorship networks respectively representing the patterns of collaboration within each decade of personality research (i.e., 1990–1999, 2000–2009, and 2010–2019). The network properties of each co‐authorship graph suggested that personality researchers have become more interconnected and collaborative with each successive decade. An examination of the semantic content of these articles suggested that the synthesis of clinical and normative trait models, along with the integration of traditional person and situation perspectives, may be driving the increased connectivity and collaboration between researchers. We hope that this novel application of network‐analytic and machine‐learning principles can serve as proof of concept for future efforts to summarize a scientific literature. [ABSTRACT FROM AUTHOR]

Nanomedicine has significant potential to revolutionize biomedicine and healthcare through innovations in diagnostics, therapeutics, and regenerative medicine. This study aims to develop a novel framework that integrates advanced natural language processing, noise-free topic modeling, and multidimensional bibliometrics to systematically identify emerging nanomedicine technology topics from scientific literature. The framework involves collecting full-text articles from PubMed Central and nanomedicine-related metrics from the Web of Science for the period 2013–2023. A fine-tuned BERT model is employed to extract key informative sentences. Noiseless Latent Dirichlet Allocation (NLDA) is applied to model interpretable topics from the cleaned corpus. Additionally, we develop and apply metrics for novelty, innovation, growth, impact, and intensity to quantify the emergence of novel technological topics. By applying this methodology to nanomedical publications, we identify an increasing emphasis on research aligned with global health priorities, particularly inflammation and biomaterial interactions in disease research. This methodology provides deeper insights through full-text analysis and leading to a more robust discovery of emerging technologies. One limitation of this study is its reliance on the existing scientific literature, which may introduce publication biases and language constraints. Additionally, manual annotation of the dataset, while thorough, is subject to subjectivity and can be time-consuming. Future research could address these limitations by incorporating more diverse data sources, and automating the annotation process. The methodology presented can be adapted to explore emerging technologies in other scientific domains. It allows for tailored assessment criteria based on specific contexts and objectives, enabling more precise analysis and decision-making in various fields. This study offers a comprehensive framework for identifying emerging technologies in nanomedicine, combining theoretical insights and practical applications. Its potential for adaptation across scientific disciplines enhances its value for future research and decision-making in technology discovery. [ABSTRACT FROM AUTHOR]

The Latent Dirichlet Location (LDA) model is a popular method for creating mixed-membership clusters. Despite having been originally developed for text analysis, LDA has been used for a wide range of other applications. We propose a new formulation for the LDA model which incorporates covariates. In this model, a negative binomial regression is embedded within LDA, enabling straight-forward interpretation of the regression coefficients and the analysis of the quantity of cluster-specific elements in each sampling units (instead of the analysis being focused on modeling the proportion of each cluster, as in Structural Topic Models). We use slice sampling within a Gibbs sampling algorithm to estimate model parameters. We rely on simulations to show how our algorithm is able to successfully retrieve the true parameter values and the ability to make predictions for the abundance matrix using the information given by the covariates. The model is illustrated using real data sets from three different areas: text-mining of Coronavirus articles, analysis of grocery shopping baskets, and ecology of tree species on Barro Colorado Island (Panama). This model allows the identification of mixed-membership clusters in discrete data and provides inference on the relationship between covariates and the abundance of these clusters. HIGHLIGHTS: We propose a new formulation for the Latent Dirichlet Allocation (LDA) model which incorporates covariates. Our extension enables a straight-forward interpretation of the regression coefficients and the analysis of the quantity of cluster-specific elements in each sampling unit - including the prediction of these quantities in a new sample through covariates. We illustrate the benefits of this formulation using three data sets: text-mining of Coronavirus articles, analysis of grocery shopping baskets, and ecology of tree species in Barro Colorado Island. We provide an R package that enables users to readily apply our model. [ABSTRACT FROM AUTHOR]

Background: Multimorbidity is common in older adults and complicates diagnosing and care for this population. Objective: We investigated co-occurrence patterns (clustering) of medical conditions in persons with Alzheimer's disease (AD) and their matched controls. Methods: The register-based Medication use and Alzheimer's disease study (MEDALZ) includes 70,718 community-dwelling persons with incident AD diagnosed during 2005-2011 in Finland and a matched comparison cohort. Latent Dirichlet Allocation was used to cluster the comorbidities (ICD-10 diagnosis codes). Modeling was performed separately for AD and control cohorts. We experimented with different numbers of clusters (also known as topics in the field of Natural Language Processing) ranging from five to 20. Results: In both cohorts, 17 of the 20 most frequent diagnoses were the same. Based on a qualitative assessment by medical experts, the cluster patterns were not affected by the number of clusters, but the best interpretability was observed in the 10-cluster model. Quantitative assessment of the optimal number of clusters by log-likelihood estimate did not imply a specific optimal number of clusters. Multidimensional scaling visualized the variability in cluster size and (dis)similarity between the clusters with more overlapping of clusters and variation in group size seen in the AD cohort. Conclusions: Early signs and symptoms of AD were more commonly clustered together in the AD cohort than in the comparison cohort. This study experimented with using natural language processing techniques for clustering patterns from an epidemiological study. From the computed clusters, it was possible to qualitatively identify multimorbidity that differentiates AD cases and controls. [ABSTRACT FROM AUTHOR]

• ACT RCTs in LMIC journals show the importance of focusing on LMIC research. • Machine learning analysis of abstracts showed diverse topics from different countries. • Research in LMICs is often nonindexed, leading to lower citations. • Ignoring non-indexed LMIC studies distorts evidence-based intervention. • Affirmative action needed from psychology professional associations and journal editors. Despite the global nature of psychological issues, an overwhelming majority of research originates from a small segment of the world's population living in high-income countries (HICs). This disparity risks distorting our understanding of psychological phenomena by underrepresenting the cultural and contextual diversity of human experience. Research from lower- and middle-income countries (LMIC) is also less frequently cited, both because it is seemingly viewed as a "special case" and because it is less well known due to language differences and biases in indexing algorithms. Acknowledging and actively addressing this imbalance is crucial for a more inclusive, diverse, and effective science of evidence-based intervention. In this state-of-the-science review, we used a machine learning method to identify key topics in LMIC research on Acceptance and Commitment Therapy (ACT), choosing ACT due to the significant body of work from LMICs. We also examined one indication of study quality (study size), and overall citations. Research in LMICs was often nonindexed, leading to lower citations, but study size could not explain a lack of indexing. Many objectively identified topics in ACT research became invisible when LMIC research was ignored. Specific countries exhibited potentially important differences in the topics. We conclude that strong and affirmative actions are needed by scientific associations and others to ensure that research from LMICs is conducted, known, indexed, and used by CBT researchers and others interested in evidence-based intervention science. [ABSTRACT FROM AUTHOR]

The advent of artificial intelligence (AI) has prompted the introduction of novel digital technologies, including mobile learning and metaverse learning, into nursing students' learning environments. This study used text network and topic modeling analyses to identify the research trends in generative AI in nursing education for students and patients in schools, hospitals, and community settings. Additionally, an ego network analysis using strengths, weaknesses, opportunities, and threats (SWOT) words was performed to develop a comprehensive understanding of factors that impact the integration of generative AI in nursing education. The literature was searched from five databases published until July 2024. After excluding studies whose abstracts were not available and removing duplicates, 139 articles were identified. The seven derived topics were labeled as usability in future scientific applications, application and integration of technology, simulation education, utility in image and text analysis, performance in exams, utility in assignments, and patient education. The ego network analysis focusing on the SWOT keywords revealed "healthcare", "use", and "risk" were common keywords. The limited emphasis on "threats", "strengths", and "weaknesses" compared to "opportunities" in the SWOT analysis indicated that these areas are relatively underexplored in nursing education. To integrate generative AI technology into education such as simulation training, teaching activities, and the development of personalized learning, it is necessary to identify relevant internal strengths and weaknesses of schools, hospitals, and communities that apply it, and plan practical application strategies aligned with clear institutional guidelines. [ABSTRACT FROM AUTHOR]

Wave and tidal energy, recognized as vital renewable resources, harness the ocean's kinetic and potential power. This study aims to provide an in-depth patent analysis of the technological landscape within these sectors. We applied a dual approach: first, a descriptive analysis was conducted to explore patent publication trends, technology lifecycle stages, patent activity by country, top assignees, and IPC classifications. Our analysis provided a detailed overview of the sector's growth and the key players involved. Second, we utilized topic modeling, specifically BERTopic enhanced with large language models, to identify and fine-tune key technological themes within the patent data. In this study, we identified seven distinct clusters each for wave and tidal energy using this approach. This method led to a novel categorization of the patents, revealing latent themes within the patent data. Although our categorization differs from traditional methods, it provides deeper insights into the thematic focus of the patents, highlighting emerging trends and areas of innovation within wave and tidal energy technologies to better exploit and optimize ocean energy conversion infrastructure. [ABSTRACT FROM AUTHOR]

The field of digital forensics relies on expertise from multiple domains, including computer science, criminology, and law. It also relies on different toolsets and an analyst's expertise to parse enormous amounts of user-generated data to find clues that help crack a case. This process of investigative analysis is often done manually. Artificial Intelligence (AI) can provide practical solutions to efficiently mine enormous amounts of data to find useful patterns that can be leveraged to investigate crimes. Natural Language Processing (NLP) is a subdomain of research under AI that deals with problems involving unstructured data, specifically language. The domain of NLP includes several tools to parse text, including topic modeling, pairwise correlation, word vector cosine distance measurement, and sentiment analysis. In this research, we propose a digital forensic investigative technique that uses an ensemble of NLP tools to identify a person of interest list based on a corpus of text. Our proposed method serves as a type of human feature reduction, where a total pool of suspects is filtered down to a short list of candidates who possess a higher correlation with the crime being investigated. [ABSTRACT FROM AUTHOR]

Interdisciplinary topic reflects the knowledge exchange and integration between different disciplines. Analyzing its evolutionary path is beneficial for interdisciplinary research in identifying potential cooperative research direction and promoting the cross-integration of different disciplines. However, current studies on the evolution of interdisciplinary topics mainly focus on identifying interdisciplinary topics at the macro level. More analysis of the evolution process of interdisciplinary topics at the micro level is still needed. This paper proposes a framework for interdisciplinary topic identification and evolutionary analysis based on BERTopic to bridge the gap. The framework consists of four steps: (1) Extract the topics from the dataset using the BERTopic model. (2) Filter out the invalid global topics and stage topics based on lexical distribution and further filter out the invalid stage topics based on topic correlation. (3) Identify interdisciplinary topics based on disciplinary diversity and disciplinary cohesion. (4) Analyze the interdisciplinary topic evolution by inspecting the intensity and content in the evolution, and visualize the evolution using Sankey diagrams. Finally, We conduct an empirical study on a dataset collected from the Web of Science (WoS) in Library & Information Science (LIS) to evaluate the validity of the framework. From the dataset, we have identified two distinct types of interdisciplinary topics in LIS. Our findings suggest that the growth points of LIS mainly exist in the interdisciplinary research topics. Additionally, our analysis reveals that more and more interdisciplinary knowledge needs to be integrated to solve more complex problems. Mature interdisciplinary topics mainly formed from the internal core knowledge in LIS stimulated by external disciplinary knowledge, while promising interdisciplinary topics are still at the stage of internalizing and absorbing the knowledge of other disciplines. The dataset, the code for implementing the algorithms, and the complete experiment results will be released on GitHub at: https://github.com/haihua0913/IITE-BERT. [ABSTRACT FROM AUTHOR]

There are several initiatives underway to improve the learning of software developers. These attempts include the integration of GitHub into software engineering classes, the creation of learning management systems, gamification approaches, and collaborative learning platforms. These initiatives have demonstrated promise in boosting students' collaborative growth and cooperation abilities, emphasizing their potential influence on improving learning experiences in practical areas. Books, on the other hand, remain basic in education, but their physical size limits their ability to explore all practical elements of a topic in depth. This limitation requires more research and application of theoretical information in real‐world circumstances. In this work, we address the issue of limited space in traditional books that frequently prevents complete presentation of practical elements of a topic. To address this issue, we propose an application that improves the reading experience and accelerates the learning process. To anticipate themes, we use a combination of latent Dirichlet allocation (LDA) algorithms and a generative pre‐trained transformer. First, utilizing LDA to find potential topic keywords inside the text and then leveraging generative pretrained transformer to predict topic names based on the LDA produced keywords. In addition, a query builder module produces and executes queries depending on the current page's topic, obtaining real‐world issues from Stack Overflow. The system classifies results by query‐title similarity, question‐answer ranking, and content quality before displaying them to users. This bridges the gap between theoretical knowledge and practical application. We illustrate the usefulness of suggested tool using simulations, comparison with existing tools and user studies. The majority of users provide favorable comments and find it interesting and helpful for improving the learning process. [ABSTRACT FROM AUTHOR]

As cognitive-inspired computation approaches, deep neural networks or deep learning (DL) models have played important roles in allowing machines to reach human-like performances in various complex cognitive tasks such as cognitive computation and sentiment analysis. This paper offers a thorough examination of the rapidly developing topic of DL-assisted aspect-based sentiment analysis (DL-ABSA), focusing on its increasing importance and implications for practice and research advancement. Leveraging bibliometric indicators, social network analysis, and topic modeling techniques, the study investigates four research questions: publication and citation trends, scientific collaborations, major themes and topics, and prospective research directions. The analysis reveals significant growth in DL-ABSA research output and impact, with notable contributions from diverse publication sources, institutions, and countries/regions. Collaborative networks between countries/regions, particularly between the USA and China, underscore global engagement in DL-ABSA research. Major themes such as syntax and structure analysis, neural networks for sequence modeling, and specific aspects and modalities in sentiment analysis emerge from the analysis, guiding future research endeavors. The study identifies prospective avenues for practitioners, emphasizing the strategic importance of syntax analysis, neural network methodologies, and domain-specific applications. Overall, this study contributes to the understanding of DL-ABSA research dynamics, providing a roadmap for practitioners and researchers to navigate the evolving landscape and drive innovations in DL-ABSA methodologies and applications. [ABSTRACT FROM AUTHOR]

With the speed increase of reviews or other forms of text, natural language has the ability to convey large and complex amounts of information in relatively small communications. This capability is being leveraged by the machine-learning algorithm known as latent dirichlet allocation (LDA), which can be utilized to discover latent topics within documents. LDA can be also used to generate summaries or abstracts from a given set of documents. However, LDA can struggle to identify topics in short documents or in data with high levels of noise. This article will introduce a new method for topic modeling with LDA based on diagonal reading for sentences (DIAG-LDA). Primarily, the features are selected using the TF-IDF algorithm, and the highest relevant features are extracted using the confidence value. Besides, the classification step is executed utilizing the LDA classifier. Ultimately, we evaluate our model using the convolutional neural network algorithm. The experiment results show that DIAG-LDA performs well in identifying features from text data, achieving a 94.4%, and 89.5% in accuracy for the datasets on international economics and the political economy. [ABSTRACT FROM AUTHOR]

This study addresses gaps in understanding accommodation satisfaction on Airbnb, focusing on Morocco as a representative of an important but unexplored region. It introduces guest count as a control variable and explores both linear and non-linear factors affecting satisfaction. Analyzing Airbnb reviews from 2013 to 2023, it emphasizes geographically-dependent determinants and variations in relevance based on host numbers. Non-linear links reveal overlooked factors, enhancing understanding beyond linear approaches. Accompanied by explanations, tables, and illustrations, the study aids Airbnb stakeholders in improving services. [ABSTRACT FROM AUTHOR]

In the present study, we investigate gender bias against politicians in a large set of news articles (n = 1,139,571) published in major media outlets in the United States between 2010 and 2020 by tracing changes in reporting about 1,095 US politicians. Using topic modeling with latent Dirichlet allocation (LDA), we identify main policy-related topics in media reports. We find gender differences in the coverage of certain policy issues, with major imbalances explained by societal factors. Specifically, we show that women in high-level political positions receive less media coverage than their male counterparts and women in less powerful positions on economic and national security issues. However, women and men in less influential positions do not differ in the amount and type of reporting they garner. Since women are still underrepresented in leadership positions, the US media may inadvertently reflect and reinforce existing gender biases in society by devoting more attention to high-profile politicians, who are overwhelmingly male. Although our longitudinal analysis shows positive changes, the gender gap in reporting continues to exist. [ABSTRACT FROM AUTHOR]

Massive Open Online Courses (MOOCs) are increasingly popular for teachers' professional development (PD). Understanding why teachers take MOOCs and how this relates to course completion could help identify underserved needs in teachers' professional learning. In the current study, we explored this question, as well as potential gaps between intention to complete the course and actual completion. Using a sample of 3,212 participants in four PD MOOCs, we applied topic modeling to open-ended and Likert-style data to identify teachers' motivations. The results show that most participants had intrinsic or professional motivations, but a subgroup of participants had prosocial motivations, namely, they wanted to support their students. In a set of logistic regression predicting course completion, we found that participants with intrinsic motivations were less likely to complete a course and participants with prosocial motivations were more likely to do so even after controlling for their initial intention. Our study contributes to the field by, first, identifying an underexplored group of learners, the prosocial learners. More research is needed to better understand this group. We also found that among teachers taking MOOCs, intrinsic motivations were associated with lower levels of engagement, contrary to findings in other populations, making a contribution to motivation theory as well as online learning practice. We concluded that the motivation-engagement relationship is more complex than previously thought, and recommend researchers continue examining this association to understand this discrepancy. Finally, we suggest practitioners take learners' a-priori motivations into account when designing MOOCs, as these could be important for course engagement. [ABSTRACT FROM AUTHOR]

Objectives: Precision Medicine (PM) uses advanced Machine Learning (ML) techniques and big data to develop personalized treatments, but healthcare still relies on traditional statistical procedures not targeted on individuals. This study investigates the impact of ML on epidemiology. Methods: A quantitative analysis of the articles in PubMed for the years 2000-2019 was conducted to investigate the use of statistical methods and ML in epidemiology. Using structural topic modelling, two groups of topics were identified and analysed over time: topics closer to the clinical side of epidemiology and topics closer to the population side. Results: The curve of the prevalence of topics associated with population epidemiology basically corresponds to the curve of the relative statistical methods, while the more dynamic curve of clinical epidemiology broadly reproduces the trend of algorithmic methods. Conclusion: The findings suggest that a renewed separation between clinical epidemiology and population epidemiology is emerging, with clinical epidemiology taking more advantage of recent developments in algorithmic techniques and moving closer to bioinformatics, whereas population epidemiology seems to be slower in this innovation. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Although trust within the entrepreneurial team is critical for its success, we have limited insights into the antecedents of a founder's trust in the team. Taking a social information processing perspective, we theorize how entrepreneurial team narratives can play an important role in building a founder's cognition‐based trust in the team. We hypothesize that the team‐level structural dimensions of diversity and distinctiveness of topics in entrepreneurial team narratives are positively related to a founder's cognition‐based trust in the team and that these relationships are less positive when the founder perceives higher levels of resource scarcity. To test our hypotheses, we apply an automated topic modeling approach to quantitatively analyze interview and survey data from 102 founders across 43 complete entrepreneurial teams. Our study has implications for research on trust in entrepreneurial teams and entrepreneurial narratives, as well as methodological implications for using topic modeling to analyze other texts in entrepreneurship research. [ABSTRACT FROM AUTHOR]