Abstract
We propose to use Twitter data as social-spatial sensors. This study deals with the question whether research papers on certain diseases are perceived by people in regions (worldwide) that are especially concerned by these diseases. Since (some) Twitter data contain location information, it is possible to spatially map the activity of Twitter users referring to certain papers (e.g., dealing with tuberculosis). The resulting maps reveal whether heavy activity on Twitter is correlated with large numbers of people having certain diseases. In this study, we focus on tuberculosis, human immunodeficiency virus (HIV), and malaria, since the World Health Organization ranks these diseases as the top three causes of death worldwide by a single infectious agent. The results of the social-spatial Twitter maps (and additionally performed regression models) reveal the usefulness of the proposed sensor approach. One receives an impression of how research papers on the diseases have been perceived by people in regions that are especially concerned by these diseases. Our study demonstrates a promising approach for using Twitter data for research evaluation purposes beyond simple counting of tweets.
Highlights
Citations have been used to measure impact of papers over decades
We focus on Twitter which is a source that has been frequently used in altmetrics research hitherto
We use the diseases human immunodeficiency virus (HIV), tuberculosis, and malaria as examples to demonstrate the use of Twitter data as social spatial sensors
Summary
Citations have been used to measure impact of papers over decades. This exclusive focus on citations in research evaluation has changed in recent years [1, 2]. Alternative metrics (altmetrics) have been proposed to measure impact broader, on academia itself. The overview of altmetrics definitions by Erdt et al [3] show that there is no formal definition of altmetrics; the definitions vary slightly. The definitions agree, that altmetrics have been proposed as an alternative or supplement to traditional bibliometrics and are based on various data sources (e.g., Twitter or Mendeley). With separate conferences (see http://www. altmetricsconference.com) and journals (see https://www.journalofaltmetrics.org), altmetrics
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
