Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 10, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000

Leveraging Large Language Models to Identify Engagement-Driving Features in Vaping-Related TikTok Videos:

Zidian Xie1, Nanda Kishore Korrapolu2, Amisha Dubey3

  • 1Clinical and Translational Science Institute, University of Rochester, 265 Crittenden Boulevard CU 420708, Rochester, NY, 14642-0708, United States, 1 5852767285.

Journal of Medical Internet Research
|November 20, 2025
PubMed
Summary

Related Concept Videos

Stimulants01:29

Stimulants

812
Stimulants are substances that enhance neural activity and elevate dopamine levels in the brain, leading to their highly addictive nature. These drugs include cocaine, amphetamines, MDMA, caffeine, and nicotine, each with distinct mechanisms of action and varied health implications.
Cocaine can be administered via snorting, injection, or smoking. It primarily functions by blocking the reuptake of dopamine, resulting in a euphoric high characterized by an intense sensation of happiness and...
812

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Understanding online sentiment toward waterpipe tobacco smoking by applying deep-learning language models to Twitter posts (2021-2023) in the United States.

Nicotine & tobacco research : official journal of the Society for Research on Nicotine and Tobacco·2026
Same author

Urgent Need to Add AI Use Measures into Nicotine and Tobacco Surveys.

Nicotine & tobacco research : official journal of the Society for Research on Nicotine and Tobacco·2026
Same author

Association between children's secondhand co-exposure to tobacco and cannabis smoke and elevated urinary cotinine levels: a cross-sectional analysis.

BMJ public health·2026
Same author

Evaluating Osteoarthritis Severity in Mice Using μCT-Derived Geometric Indices.

Biology·2026
Same author

Characterising 'Spree Bar': an examination of the popularity, marketing and composition of a vaping device with a nicotine analogue across multiple data streams.

Tobacco control·2026
Same author

Machine learning approaches for crystallographic classification from synthetic 2D X-ray diffraction data.

Journal of applied crystallography·2026
This summary is machine-generated.

Young adults engage more with TikTok videos about vaping when they feature cars, talking, emojis, or humor, not promotions. These insights can guide vaping prevention campaigns on social media.

Area of Science:

  • Social Media Analysis
  • Public Health Communication
  • Digital Media Research

Background:

  • Electronic cigarette (e-cigarette) use is widespread among young people in the United States.
  • TikTok is a primary platform for e-cigarette content, often featuring promotional material.

Purpose of the Study:

  • To identify key features of e-cigarette TikTok videos that drive high user engagement.
  • To inform the design of future vaping prevention campaigns on social media.

Main Methods:

  • Analyzed 1487 e-cigarette TikTok videos from January 2023 to January 2024 using the TikTok API.
  • Employed GPT-4 and Video-LLaMA large language models for video feature extraction.
  • Utilized a linear mixed effects model to identify features associated with user engagement (likes+shares+comments/views).
Keywords:
AITikTokartificial intelligencee-cigaretteelectronic cigarettespreventionsocial mediauser engagementvaping

Related Experiment Videos

Last Updated: Jan 10, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000

Main Results:

  • GPT-4 demonstrated higher accuracy (83%-100%) than Video-LLaMA (24%-88%) in feature identification.
  • Videos set in cars (RR=3.91), featuring young adults (RR=1.24), talking (RR=1.63), emojis (RR=1.88), or humor (RR=1.61) significantly increased engagement.
  • Promotional e-cigarette content decreased engagement (RR=0.40).

Conclusions:

  • Specific TikTok video elements, including background settings, presenter demographics, and content style, significantly influence user engagement.
  • Findings provide actionable insights for creating effective vaping prevention content on social media platforms like TikTok.