AU - JOUR AU - Bian, AU江，AU云鹏，AU - Salloum, Ramzi G - Guo, AU Yi, AU - Wang, Mo AU - Prosperi, Mattia AU - Zhang, Hansi AU - Du, Xinsong AU - Ramirez-Diaz, Laura J AU - He, AU哲，Sun, Yuan PY - 2017da - 2017/12/13 TI -利用社交媒体数据了解促销信息对外行讨论的影响:社交媒体林奇综合征背景:社交媒体正被制药公司、政府机构、卫生保健组织、专业人士和新闻媒体等各种利益相关者使用，作为吸引受众提高疾病意识并最终改善公众健康的一种方式。然而，目前还不清楚这些健康信息对外行人有什么影响。目的:本研究旨在详细考察与Lynch综合征相关的健康宣传信息如何影响外行人在社交媒体平台(Twitter)上的话题意识和态度。方法:利用话题建模和情感分析技术对Lynch综合征相关推文进行研究，回答以下研究问题(RQs):(1) Lynch综合征相关推文中讨论最多的话题是什么?(2) Twitter上与Lynch综合征相关的宣传信息如何影响外行人的讨论?(3)大肠癌宣传月和Lynch综合征宣传日的Lynch综合征宣传活动对外行人的讨论和态度有何影响?特别地，我们使用一组关键词，通过Twitter公共搜索应用程序编程接口(API)收集2016年10月26日至2017年8月11日(289天)与Lynch综合征相关的推文。我们尝试了两种不同的分类方法，将tweet分为以下三类:(1)不相关的，(2)促销健康信息，(3)外行人的讨论。我们采用主题建模的方法来发现这些与Lynch综合征相关的推文的主题，并对每条外行人的推文进行情绪分析，以衡量作者对Lynch综合征的态度(即积极、消极和中立)。 The topic modeling and sentiment analysis results were elaborated to answer the three RQs. Results: Of all tweets (N=16,667), 87.38% (14,564/16,667) were related to Lynch syndrome. Of the Lynch syndrome–related tweets, 81.43% (11,860/14,564) were classified as promotional and 18.57% (2704/14,564) were classified as laypeople’s discussions. The most discussed themes were treatment (n=4080) and genetic testing (n=3073). We found that the topic distributions in laypeople’s discussions were similar to the distributions in promotional Lynch syndrome–related information. Furthermore, most people had a positive attitude when discussing Lynch syndrome. The proportion of negative tweets was 3.51%. Within each topic, treatment (16.67%) and genetic testing (5.60%) had more negative tweets compared with other topics. When comparing monthly trends, laypeople’s discussions had a strong correlation with promotional Lynch syndrome–related information on awareness (r=.98, P<.001), while there were moderate correlations on screening (r=.602, P=.05), genetic testing (r=.624, P=.04), treatment (r=.69, P=.02), and risk (r=.66, P=.03). We also discovered that the Colon Cancer Awareness Month (March 2017) and the Lynch Syndrome Awareness Day (March 22, 2017) had significant positive impacts on laypeople’s discussions and their attitudes. Conclusions: There is evidence that participative social media platforms, namely Twitter, offer unique opportunities to inform cancer communication surveillance and to explore the mechanisms by which these new communication media affect individual health behavior and population health. SN - 1438-8871 UR - //www.mybigtv.com/2017/12/e414/ UR - https://doi.org/10.2196/jmir.9266 UR - http://www.ncbi.nlm.nih.gov/pubmed/29237586 DO - 10.2196/jmir.9266 ID - info:doi/10.2196/jmir.9266 ER -