We analyze the information credibility of news propagated through Twitter, a popular microblogging service. Previous research has shown that most of the messages posted on Twitter are truthful, but the service is also used to spread misinformation and false rumors, often unintentionally.On this paper we focus on automatic methods for assessing the credibility of a given set of tweets. Specifically, we analyze microblog postings related to "trending" topics, and classify them as credible or not credible, based on features extracted from them. We use features from users' posting and re-posting ("re-tweeting") behavior, from the text of the posts, and from citations to external sources.We evaluate our methods using a significant number of human assessments about the credibility of items on a recent sample of Twitter postings. Our results shows that there are measurable differences in the way messages propagate, that can be used to classify them automatically as credible or not credible, with precision and recall in the range of 70% to 80%.

/pdf/information-credibility-on-twitter-2d1l7ckae1.pdf

Information credibility on twitter

We analyze microblog posts generated during two recent, concurrent emergency events in North America via Twitter, a popular microblogging service. We focus on communications broadcast by people who were "on the ground" during the Oklahoma Grassfires of April 2009 and the Red River Floods that occurred in March and April 2009, and identify information that may contribute to enhancing situational awareness (SA). This work aims to inform next steps for extracting useful, relevant information during emergencies using information extraction (IE) techniques.

/pdf/microblogging-during-two-natural-hazards-events-what-twitter-9brw09o2kj.pdf

Microblogging during two natural hazards events: what twitter may contribute to situational awareness

Background
Surveys are popular methods to measure public perceptions in emergencies but can be costly and time consuming. We suggest and evaluate a complementary “infoveillance” approach using Twitter during the 2009 H1N1 pandemic. Our study aimed to: 1) monitor the use of the terms “H1N1” versus “swine flu” over time; 2) conduct a content analysis of “tweets”; and 3) validate Twitter as a real-time content, sentiment, and public attention trend-tracking tool.

/pdf/pandemics-in-the-age-of-twitter-content-analysis-of-tweets-rp5bypn75b.pdf

Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak

Background: Twitter is an instant micro-blogging service that allows users to post, read, and exchange information and thoughts easily with masses across the globe. In response to the 2009 Influenza A virus subtype H1N1 outbreak (aka "swine flu"), users produced thousands of posts on the subject. Media outlets have claimed that Twitter and other forms of social media have led to the viral distribution of mass misinformation and may be a threat to public health and government initiatives. However, quantifiable evidence of these claims has not been substantiated. 

Objective: This exploratory project aims to analyze the content of Twitter posts or “tweets” shared during the H1N1 outbreak to determine the types and quality of information that social media users are exchanging in pandemics.

Methods: Using the Infovigil system, an emerging infoveillance system, we are continuously identifying and archiving health-related tweets. Between April 28 and May 11, 2009, we archived over 300,000 tweets containing the keywords or hashtags “swine flu”, “swineflu”, or “H1N1”. A random selection of tweets from each hour of each day were coded for content by two raters. A multi-axial coding scheme was created using an iterative process to reflect the range of data. Data analysis consisted of descriptive statistics and univariate analysis of content between days. Non-English posts and reposts (“retweets”) were excluded from the analysis.

Results: Preliminary analysis of 400 tweets indicates that news posts were the most common type of information shared (46%) followed by public health education (19.18%) and H1N1-related humour (18.25%). 36.75% of all posts quoted news articles verbatim and provided URLs to the source. Only 7 cases could be identified as possible sources of misinformation. 

Conclusions: Contrary to anecdotal evidence, misinformation is not rampantly spread via Twitter. Instead, the service is being utilized to distribute news and information from credible sources and almost one of five tweets are of humorous nature. Contrary to some media reports of Twitter fueling an epidemic of misinformation, Twitter can and is already used to quickly disseminate pandemic information to the public. Further analysis of tweets collected during an epidemic will allow us to refine the Infovigil system for twitter-based syndromic surveillance []

Pandemics in the Age of Twitter: Content Analysis of “Tweets” During the H1N1 Outbreak

In this article we explore the behavior of Twitter users under an emergency situation. In particular, we analyze the activity related to the 2010 earthquake in Chile and characterize Twitter in the hours and days following this disaster. Furthermore, we perform a preliminary study of certain social phenomenons, such as the dissemination of false rumors and confirmed news. We analyze how this information propagated through the Twitter network, with the purpose of assessing the reliability of Twitter as an information source under extreme circumstances. Our analysis shows that the propagation of tweets that correspond to rumors differs from tweets that spread news because rumors tend to be questioned more than news by the Twitter community. This result shows that it is posible to detect rumors by using aggregate analysis on tweets.

Twitter under crisis: can we trust what we RT?

This paper offers a descriptive account of Twitter (a microblogging service) across four high-profile, mass convergence events - two emergency and two national security. We statistically examine how Twitter is being used surrounding these events, and compare and contrast how that behaviour is different from more general Twitter use. Our findings suggest that Twitter messages sent during these types of events contain more displays of information broadcasting and brokerage, and we observe that general Twitter use seems to have evolved over time to offer more of an information-sharing purpose. We also provide preliminary evidence that Twitter users who join during and in apparent relation to a mass convergence or emergency event are more likely to become long-term adopters of the technology.

Twitter adoption and use in mass convergence and emergency events

This paper considers a subset of the computer-mediated communication (CMC) that took place during the flooding of the Red River Valley in the US and Canada in March and April 2009. Focusing on the use of Twitter, a microblogging service, we identified mechanisms of information production, distribution, and organization. The Red River event resulted in a rapid generation of Twitter communications by numerous sources using a variety of communications forms, including autobiographical and mainstream media reporting, among other types. We examine the social life of microblogged information, identifying generative, synthetic, derivative and innovative properties that sustain the broader system of interaction. The landscape of Twitter is such that the production of new information is supported through derivative activities of directing, relaying, synthesizing, and redistributing, and is additionally complemented by socio-technical innovation. These activities comprise self-organization of information.

Chatter on the red: what hazards threat reveals about the social life of microblogged information

Crises and disasters have micro and macro social arrangements that differ from routine situations, as the field of disaster studies has described over its 100-year history. With increasingly pervasive information and communications technology and a changing political arena where terrorism is perceived as a major threat, the attention to crisis is high. Some of these new features of social life have created changes in disaster response that we are only beginning to understand. The University of Colorado is establishing an area of sociologically informed research and information and communications technology development in crisis informatics. This article reports on research that examines features of computer-mediated communication and information sharing activity during and after the April 16, 2007, crisis at Virginia Tech by members of the public. The authors consider consequences that these technology-supported social interactions have on emergency response and implications for methods in e-Social Science.

Crisis in a Networked World

Social media and other online communication tools are a subject of great interest in mass emergency response. Members of the public are turning to these solutions to seek and offer emergency information. Emergency responders are working to determine what social media policies should be in terms of their "public information" functions. We report on the online communications from all the coastal fire and police departments within a 100 mile radius of Hurricane Sandy's US landfall. Across four types of online communication media, we collected data from 840 fire and police departments. Findings indicate that few departments used these online channels in their Sandy response efforts, and that communications differed between fire and police departments and across media type. However, among the highly engaged departments, there is evidence that they bend and adapt policies about what constitutes appropriate public communication in the face of emergency demands; therefore, we propose that flexibility is important in considering future emergency online communication policy. We conclude with design recommendations for making online communication media more "listenable" for both emergency managers and members of the public.

Amanda Lee Hughes

Papers

Microblogging during two natural hazards events: what twitter may contribute to situational awareness

Twitter adoption and use in mass convergence and emergency events

Chatter on the red: what hazards threat reveals about the social life of microblogged information

Crisis in a Networked World

Online public communications by police & fire services during the 2012 Hurricane Sandy