DOI: 10.1007/s11069-020-04024-6
论文题名: Dietary pattern recognition on Twitter: a case example of before, during, and after four natural disasters
作者: Turner-McGrievy G. ; Karami A. ; Monroe C. ; Brandt H.M.
刊名: Natural Hazards
ISSN: 0921030X
出版年: 2020
卷: 103, 期: 1 起始页码: 1035
结束页码: 1049
语种: 英语
中文关键词: Diet
; Diet patterns
; Natural disaster
; Social media
; Twitter
英文关键词: diet
; disaster management
; food availability
; food quality
; food supply
; hurricane event
; information and communication technology
; natural disaster
; pattern recognition
; quantitative analysis
; social media
; South Carolina
; United States
英文摘要: Little is known about what foods/beverages (F&B) are common during natural disasters. The goal of this study was to track high-frequency F&B mentions during four hurricanes affecting the coast of South Carolina for quantifying dietary patterns in Twitter. A listing of common F&B (n = 173) was created from the top food sources of energy, fat, protein, and carbohydrate in the USA. A sampling of > 500,000 tweets containing hashtag names (e.g., #HurricaneFlorence) or actual names (e.g., “Hurricane Florence”) of the four hurricanes was collected using Crimson Hexagon. ANOVA was used to examine differences in number of mentions in each food group pre- (6 days before), during (48 h of the hurricane), and post-hurricane (6 days after). Descriptive statistics were used to examine the most frequently mentioned F&B (threshold defined as ≥ 4 mentions/day for each F&B item or 10% of the foods mentioned) and whether F&B were top sources of energy/macronutrients. More than 5000 mentions of F&B were collected in our sample. Grains were the most frequently mentioned food group pre-hurricane, and dairy was most frequently mentioned during the hurricanes. The top five most commonly mentioned F&B overall were milk (n = 517), pizza (n = 511), turkey (n = 425), oranges (n = 384), and waffles (n = 346). Foods mentioned were commonly energy and protein dense. Five foods (pizza, waffles, milk, rolls, and bread) were categorized as a top contributor across energy and all three macronutrients. Social media may be a unique way to detect dietary patterns and help inform public health social media campaigns to advise people about stocking up on healthy, non-perishable foods ahead of natural disasters. © 2020, Springer Nature B.V. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/168771
Appears in Collections: 气候变化与战略
There are no files associated with this item.
作者单位: Arnold School of Public Health, University of South Carolina, 915 Greene Street, Columbia, SC 29208, United States; School of Library and Information Science, University of South Carolina, 1501 Greene Street, Columbia, SC 29208, United States
Recommended Citation:
Turner-McGrievy G.,Karami A.,Monroe C.,et al. Dietary pattern recognition on Twitter: a case example of before, during, and after four natural disasters[J]. Natural Hazards,2020-01-01,103(1)