A Topic Modeling Comparison Between LDA, NMF, Top2Vec, and BERTopic to Demystify Twitter Posts

Frontiers in Sociology, Journal Year: 2022, Volume and Issue: 7

Published: May 6, 2022

The richness of social media data has opened a new avenue for science research to gain insights into human behaviors and experiences. In particular, emerging data-driven approaches relying on topic models provide entirely perspectives interpreting phenomena. However, the short, text-heavy, unstructured nature content often leads methodological challenges in both collection analysis. order bridge developing field computational empirical research, this study aims evaluate performance four modeling techniques; namely latent Dirichlet allocation (LDA), non-negative matrix factorization (NMF), Top2Vec, BERTopic. view interplay between relations digital media, takes Twitter posts as reference point assesses different algorithms concerning their strengths weaknesses context. Based certain details during analytical procedures quality issues, sheds light efficacy using BERTopic NMF analyze data.

Language: Английский

---

Trust me, I'm an influencer! - Causal recipes for customer trust in artificial intelligence influencers in the retail industry

Journal of Retailing and Consumer Services, Journal Year: 2022, Volume and Issue: 72, P. 103242 - 103242

Published: Dec. 30, 2022

Language: Английский

---

All hype or the real deal? Investigating user engagement with virtual influencers in tourism

Tourism Management, Journal Year: 2023, Volume and Issue: 99, P. 104779 - 104779

Published: May 9, 2023

Virtual influencer (VI) marketing has become increasingly prevalent on social media with the evolving landscape of digitalisation. However, use VIs in tourism received limited attention literature. This study examines salient source and content attributes that stimulate Instagram users' engagement a context. An online discrete choice experiment was designed using key (i.e. realness, image composition caption discourse) identified from focus groups. Survey responses 309 adult users Australia were analysed through modelling. The findings indicate humanlike are preferred over 3D animated least influencers 2D VIs. posts combine images settings rational messages attract most audience. Theoretical practical implications provided recommendations for how practitioners can harness VI effectively.

Language: Английский

---

Unraveling the power of social media influencers: Qualitative insights into the role of Instagram influencers in the hospitality and tourism industry

Journal of Hospitality and Tourism Management, Journal Year: 2024, Volume and Issue: 58, P. 214 - 243

Published: Feb. 10, 2024

Language: Английский

---

Past, present, and future scene of influencer marketing in hospitality and tourism management

Journal of Travel & Tourism Marketing, Journal Year: 2024, Volume and Issue: 41(3), P. 322 - 343

Published: Feb. 19, 2024

This paper reviews influencer marketing (IM)-research in the tourism and hospitality (H&T) field to provide a state-of-the-art of past, present, future IM-research. It identifies research themes, theoretical underpinnings, methodologies utilized Seventy-five studies are examined through co-citation, bibliographic coupling content analysis based on TCCM (theory/context/characteristics/methodology) framework. Co-citation reveals that IM-research focused travel blogs, vlogs, celebrity endorsement, authenticity. An assessment recent indicates partial shift toward study more contemporary issues. field-discipline-focused hybrid-review contributes literature practice by presenting

Language: Английский

---

What influences tourist behaviors during and after the COVID-19 pandemic? Focusing on theories of risk, coping, and resilience

Journal of Hospitality and Tourism Management, Journal Year: 2022, Volume and Issue: 50, P. 355 - 365

Published: Feb. 11, 2022

Language: Английский

---

From interaction to relationship: Rethinking parasocial phenomena in travel live streaming

Tourism Management, Journal Year: 2022, Volume and Issue: 93, P. 104583 - 104583

Published: June 14, 2022

Language: Английский

---

Risk, Trust, and the Roles of Human Versus Virtual Influencers

Journal of Travel Research, Journal Year: 2023, Volume and Issue: 63(6), P. 1370 - 1394

Published: Aug. 19, 2023

Drawing on the theory of sociology trust and risk, this study proposes empirically tests a conceptual model tourists’ new destination visit intentions. The links subjective knowledge to risk perceptions explores moderating effects social media influencers (human vs. virtual) psychographic factors in context. Data were collected from two studies: Study 1, through survey distributed participants Malaysia ( n = 493 valid responses); 2, between-subjects design experiment with another sample 470 responses). findings expand tourism research by showing that mediates how influences perceived risk. Optimism life satisfaction have significant Furthermore, relationship between intention is moderated human virtual influencers, according whether their message positive or negative.

Language: Английский

---

Progress on image analytics: Implications for tourism and hospitality research

Tourism Management, Journal Year: 2023, Volume and Issue: 100, P. 104798 - 104798

Published: June 10, 2023

Despite the increased recognition of importance images as a rich data source, full potential to advance tourism and hospitality knowledge, both conceptually theoretically, has yet be fully tapped. This study provides critical review image analytics, examining its broad implications for research. paper offers progress definitions, features related theories, well presents methodological framework conducting image-related studies, complementing dominant textual analysis used in The makes contributions literature by developing point reference application analytics studies.

Language: Английский

---

The role of social media marketing activities in driving self–brand connection and user engagement behavior on Instagram: a moderation–mediation approach

European Journal of Innovation Management, Journal Year: 2023, Volume and Issue: 27(5), P. 1723 - 1742

Published: Jan. 21, 2023

Purpose This study explores central questions related to the connection between social media marketing activities (SMMAs), user engagement and self-brand of restaurant Instagram pages. The examines mediating role SMMAs self–brand connections. Also, this through moderating gender trust. Design/methodology/approach A convenience sample method was employed collect data from customers (18–24 years old). structural equation modeling approach PROCESS macro were applied based on 298 online questionnaires completed by who follow effect for trust performed. Findings findings revealed that have a significant positive influence engagement. Further, acts as mediator connection. results illustrate importance in enhancing light Practical implications paper presents managerial businesses about how can effectively enhance behavior Originality/value research developed theoretical model understand might industry invoking variables relationship offers new practical contributions add value (SMM) literature testing moderated–mediation these constructs hospitality sector.

Language: Английский

---