Skip to Content


You searched for: Keyword Social media Remove constraint Keyword: Social media

Search Results

  1. Recognizing Images of Eating Disorders in Social Media [Download]

    Title: Recognizing Images of Eating Disorders in Social Media
    Author: Counts, Samsara
    Description: Eating disorders (ED) are pervasive and do not discriminate based on race, religion, gender, or socioeconomic status. Comorbidities include anxiety, depression, substance abuse, self injurious behaviors, and history of trauma. ED are often a lifelong struggle, with approximately ⅔ of patients never achieving a full and sustained remission. Exposure to media expressing “the thin ideal” can be triggering to individuals with ED as well as those at risk for developing them. Social media is rife with these triggers. Concurrent with the rise of social media, individuals with ED have created communities in which they support one another in the dangerous pursuit of this illness' goal: to be “thin enough.” Websites promoting anorexia (pro-ana) and ED as lifestyle choices valorize acting on ED symptoms. Such sites teach those suffering or at risk from ED how to act on the illness and support them in doing so, putting them at risk for severe health complications. The impact of images in this community far exceeds that of other communities surrounding physical and mental health issues. Therefore, it is essential that clinicians and family members be able to identify websites containing images associated with the promotion of ED to prevent exposure to these triggers. This research aims to automatically detect such triggering material, with the ultimate goal of designing parental and clinical controls. We report on a proof of concept, machine learning approach to identify pro-ana content, trained on example data from online social media searches. The training data was chosen to compare pro-ana content with other content similar in demographics and photographic style, composed of the hashtag-based categories #proana, #selfie, #ootd, and #greek. We randomly chose 20% of these images as test data and train the Resnet Deep Learning neural network to classify the remaining images. On test data this gives 81% classification accuracy—a significant improvement over chance (25%). These proof of concept results suggest that it is feasible to automatically detect social media sources with triggering material, informing the creation of tools that can assist clinicians and family members to improve health outcomes. We used the classifier to make a web application that assesses how pro-ana a social media user’s content is. The tool, designed for clinicians, allows them to enter a social media username and then gives an analysis of that user’s online presence, classifying its content. The tool also displays a hashtag similarity map showing trending hashtags closely related to #proana.
    Keywords: Research Days 2018, Social media, Eating disorders
    Date Uploaded: 04/21/2018
  2. Culture, Time-Orientation, Coping Styles and their Effects on Procrastination [Download]

    Title: Culture, Time-Orientation, Coping Styles and their Effects on Procrastination
    Author: Shen, Ruihang
    Description: With the rapid development of new media in the present age, procrastination has become increasingly prevalent, especially among students. With considerable negative consequences on physical and mental health, academic and career achievements and financial and relationship aspects, a sizable body of research have examined various factors that influence the extent to which individuals procrastinate. However, most current research studying procrastination focuses on western, English-speaking countries. Also, though some research identifies time-orientation can be a significant predictor of procrastination, few study connect culture influence with time-orientation. Building on other studies, this project seeks to understand whether individuals’ time-orientations and copy styles mediate the influence of culture on procrastination. Theoretically, this study will fill in the gap of the previous study and extend people’s understanding of procrastination. Data are being collected from undergraduate students at the George Washington University. Participants will be approximately 75 domestic American students and 75 sojourning students originally from China. Participants will be asked to fill out a survey questionnaire that measures people’s considerations for future consequences, coping styles, motives of social media use, and tendency to procrastinate. Collected data will be analyzed using statistical analysis techniques, such as multiple regression. Results from the research will support or reject the following hypotheses: H1: Chinese students will have greater concern for future consequences than American students. H2: Concern for future consequence will be negatively associated with the use of social media to escape/relax, and will be positively associated with the use of social media to learn/get help. H3: The use of social media to escape/relax is positively associated with the tendency to procrastinate, whereas the use of social media to learn/get help is negatively associated with the tendency to procrastinate. H4: Concern for future consequences and motives for social media use will mediate the effect of culture on the tendency to procrastinate. The research will deepen people’s understanding of the role of culture and social media use in shaping individuals’ tendencies to procrastinate, thus helping people, especially college students, to control their procrastination tendencies.
    Keywords: Research Days 2018, Mental health, Cultural influence, Social media
    Date Uploaded: 04/14/2018
  3. The Provenance of a Tweet [Download]

    Title: The Provenance of a Tweet
    Author: Kerchner, Daniel
    Description: The purpose of this paper is to describe the metadata the SFM team has chosen to record about the source of the social media items SFM collects, with the aim of advancing the conversation around social media data.
    Keywords: Social media, Twitter, Provenance, Research methodology
    Date Uploaded: 06/20/2016