Exploratory Analysis of Marketing and Non-marketing E-cigarette Themes on Twitter.

Exploratory Analysis of Marketing and Non-marketing E-cigarette Themes on Twitter.
复制标题

DOI:
10.1007/978-3-319-47874-6_22
复制
发表时间:
2016-11
期刊:
Social informatics : 8th International Conference, SocInfo 2016, Bellevue, WA, USA, November 11-14, 2016, Proceedings. Part II. SocInfo (Conference) (8th : 2016 : Bellevue, Wash.)
影响因子:
--
通讯作者:
Kavuluru R
Kavuluru R
中科院分区:
其他
文献类型:
--
作者:
Han S;Kavuluru R

文献摘要

参考文献

被引文献

相似文献

电子烟 (e-cigs) 自从 2007 年在美国推出以来,一直越来越受欢迎,并成为一种有争议的烟草产品。电子烟的无烟特性使其比传统香烟的危害更小,也是计划戒烟的人使用电子烟的主要原因之一。美国食品和药物管理局 (FDA) 于 2016 年 5 月上旬推出了新法规,并于 2016 年 8 月 8 日生效。鉴于这一重要背景,在本文中,我们报告了一个项目的结果,该项目根据主题建模生成的主题的语义解释来识别电子烟推文中的当前主题。鉴于营销/广告推文几乎占所有电子烟推文的一半,我们首先构建一个分类器,根据 1000 条推文的手工数据集识别营销和非营销推文。将分类器应用于超过一百万条推文的数据集(在 4/2015 – 6/2016 期间收集)后,我们进行初步的内容分析,并在使用主题连贯性识别适当数量的主题后,分别对两组推文运行主题模型。我们通过将生成的主题与特定的电子烟主题相关联来解释主题建模过程的结果。我们还使用 GeoNames API 报告从在特定地点(例如学校和教堂)生成的电子烟推文中识别出的主题,以获取在我们的数据集中找到的地理标记推文。据我们所知,这是首次采用主题建模来识别一般电子烟主题,并在与特定兴趣地点相关的地理标记推文的背景下进行识别。
Electronic cigarettes (e-cigs) have been gaining popularity and have emerged as a controversial tobacco product since their introduction in 2007 in the U.S. The smoke-free aspect of e-cigs renders them less harmful than conventional cigarettes and is one of the main reasons for their use by people who plan to quit smoking. The US food and drug administration (FDA) has introduced new regulations early May 2016 that went into effect on August 8, 2016. Given this important context, in this paper, we report results of a project to identify current themes in e-cig tweets in terms of semantic interpretations of topics generated with topic modeling. Given marketing/advertising tweets constitute almost half of all e-cig tweets, we first build a classifier that identifies marketing and non-marketing tweets based on a hand-built dataset of 1000 tweets. After applying the classifier to a dataset of over a million tweets (collected during 4/2015 – 6/2016), we conduct a preliminary content analysis and run topic models on the two sets of tweets separately after identifying the appropriate numbers of topics using topic coherence. We interpret the results of the topic modeling process by relating topics generated to specific e-cig themes. We also report on themes identified from e-cig tweets generated at particular places (such as schools and churches) for geo-tagged tweets found in our dataset using the GeoNames API. To our knowledge, this is the first effort that employs topic modeling to identify e-cig themes in general and in the context of geo-tagged tweets tied to specific places of interest.
DOI: 10.2196/jmir.4466
发表时间: 2015-11-06
影响因子: 7.4
作者:
Kim AE;Hopper T;Simpson S;Nonnemaker J;Lieberman AJ;Hansen H;Guillory J;Porter L
通讯作者: Porter L
DOI: 10.1136/tobaccocontrol-2014-051551
发表时间: 2014-07
期刊: Tobacco control
影响因子: 5.2
作者:
Huang J;Kornfield R;Szczypka G;Emery SL
通讯作者: Emery SL
DOI: 10.1152/ajplung.00170.2016
发表时间: 2016-07-01
影响因子: 4.9
作者:
Martin, Elizabeth M.;Clapp, Phillip W.;Jaspers, Ilona
通讯作者: Jaspers, Ilona
DOI: 10.1136/tc.2010.042168
发表时间: 2011-05
期刊: Tobacco control
影响因子: 5.2
作者:
Etter JF;Bullen C;Flouris AD;Laugesen M;Eissenberg T
通讯作者: Eissenberg T
社交聆听: Twitter 上电子烟讨论的内容分析。
DOI: 10.2196/jmir.4969
发表时间: 2015-10-27
影响因子: 7.4
作者:
Cole-Lewis H;Pugatch J;Sanders A;Varghese A;Posada S;Yun C;Schwarz M;Augustson E
通讯作者: Augustson E