Abbasi, A., Adjeroh, D., Dredze, M., Paul, M. J., Zahedi, F. M., Zhao, H., Ross, A. (2014). Social Media Analytics for Smart Health. IEEE Intelligent Systems, 29(2), 60—80. doi:10.1109/MIS.2014.29
People increasingly use social media for obtaining health-related information. Abbasi and Adjeroh discuss the following five dimensions to judge a medium's importance as an information provider:
- credibility; an analysis of health-related website which used data taken from third-party health URL databases showed that credible sites are closely connected, while less-credible sites are dispersed across various clusters.
- recency was measured with time series data on product recalls, ongoing reviews, and drug safety communications which showed that for roughly 35% of the events social media was providing leading signals, contrasted to lagging signals for 50% of the observed events.
- uniqueness indicated that especially leading signals obtained from social media are often unique,
- frequency which is in general inversely proportional to salience, and
- salience, indicating the extend to which a topic is discussed.