Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4937545 | Computers in Human Behavior | 2017 | 26 Pages |
Abstract
Searching for specific topics on Twitter, readers have to judge the credibility of tweets. In this paper, we examine the relationship between reader demographics, news attributes and tweet features with reader's credibility perception, and further examine the correlation among these factors. We found that reader's educational background and geo-location have significant correlation with their credibility perception, and furthermore the news attributes in tweets are also significantly correlated with reader's credibility perception. Despite differences in demographics, readers find features including the search topic keyword and the writing style of tweets most helpful in perceiving tweet credibility. While previous studies reported the use of specific features, our results showed that readers use combination of features to make decisions regarding tweet credibility. Comparing the credibility level predicted by an automatic prediction tool and that by reader's perception, we found that readers tend to be more trusting, possibly due to the limited explicit author information available on Twitter. Our study can help devise strategies to enhance the tweet credibility with readers and also help educate readers to be more cautious with information credibility on Twitter.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Shafiza Mohd Shariff, Xiuzhen Zhang, Mark Sanderson,