Understanding e-cigarette content and promotion on YouTube through machine learning.
Understanding e-cigarette content and promotion on YouTube through machine learning.
复制标题
DOI：
10.1136/tobaccocontrol-2021-057243
复制
发表时间：
2023-11
期刊：
影响因子：
5.2
通讯作者：
Murthy, Dhiraj
中科院分区：
文献类型：
--
作者：
Kong, Grace;Schott, Alex Sebastian;Lee, Juhan;Dashtian, Hassan;Murthy, Dhiraj
关键词：
YouTube是年轻人使用的流行社交媒体，有电子烟内容。我们使用机器学习来识别电子烟视频的内容、特色电子烟产品、视频上传者以及电子烟产品的营销和销售。我们使用18个搜索词（例如，e-cig），并使用元数据作为监督机器学习的输入预测四个模型：1）视频主题，2）特色电子烟产品，3）频道类型（即，视频上传者），以及4）折扣/销售。我们评估了参与数据与四种模型之间的关联。3830个英语视频被纳入监督机器学习。最常见的视频主题是“产品评论”（48.9%），其次是“说明”（例如，“如何”使用/修改电子烟; 17.3%）;展示了多样化的电子烟产品;“vape爱好者”最常发布电子烟视频（53.1%），其次是零售商（19.0%）; 43.2%的视频有电子烟的折扣/销售;最常见的销售策略是购买外部链接（31.5%）。“电子烟把戏”是最不常见的主题，但参与度最高（例如，> 200万次观看）。“大麻”（53.9%）和“说明”（49.9%）主题更有可能有外部链接用于购买（p<.001）。这4个模型的F1得分（衡量模型准确性的指标）高达0.87。我们的研究结果表明，在年轻人可以访问的YouTube视频中，各种电子烟产品通过不同的视频主题进行展示，并提供折扣/销售。调查结果凸显了监管社交媒体平台上电子烟推广的必要性。
YouTube is a popular social media used by youth and has e-cigarette content. We used machine learning to identify the content of e-cigarette videos, featured e-cigarette products, video uploaders, and marketing and sales of e-cigarette products. We identified e-cigarette content using 18 search terms (e.g., e-cig) using fictitious youth viewer profiles and predicted four models using the metadata as the input to supervised machine learning: 1) video themes, 2) featured e-cigarette products, 3) Channel type (i.e., video uploaders), and 4) discount/sales. We assessed the association between engagement data and the four models. 3830 English videos were included in the supervised machine learning. The most common video theme was “product review” (48.9%) followed by “instruction” (e.g., “how to” use/modify e-cigarettes; 17.3%); diverse e-cigarette products were featured; “vape enthusiasts” most frequently posted e-cigarette videos (53.1%) followed by retailers (19.0%); 43.2% of videos had discount/sales of e-cigarettes; and the most common sales strategy was external links for purchasing (31.5%). “Vape trick” was the least common theme but had the highest engagement (e.g., >2 million views). “Cannabis” (53.9%) and “instruction” (49.9%) themes were more likely to have external links for purchasing (p<.001). The 4 models achieved an F1 score (a measure of model accuracy) of up to .87. Our findings indicate that on YouTube videos accessible to youth, a variety of e-cigarette products are featured through diverse videos themes, with discount/sales. The findings highlight the need to regulate the promotion of e-cigarettes on social media platforms.
登录
查看更多内容
影响因子：
5.2
作者：
Huang J;Kornfield R;Szczypka G;Emery SL
通讯作者：
Emery SL
影响因子：
4.5
作者：
Luo C;Zheng X;Zeng DD;Leischow S
通讯作者：
Leischow S
影响因子：
7.4
作者：
Visweswaran，Shyam;Colditz，Jason B.;Primack，Brian A.
通讯作者：
Primack，Brian A.
影响因子：
12.7
作者：
Cheney, Marshall;Gowin, Mary;Wann, Taylor Franklin
通讯作者：
Wann, Taylor Franklin
影响因子：
4.2
作者：
Lee, Juhan;Kong, Grace;Salloum, Ramzi G.
通讯作者：
Salloum, Ramzi G.