Sentiment analysis in social networks of health institutions

Abstract

Twitter, a communication platform that creates a social impact; it conveys the messages of non-profit organizations to the masses and the emotions of the masses to non-profit organizations. This research; It aims to examine Twitter posts about health-related non-profit organizations, to determine the emotional states about these institutions on social media and to measure these feelings. Sentiment analysis about WHO, ILO, IOM, UNICEF, FAO, Red Cross, UNDP and UNHCR were carried out using the R program. The tweets used in sentiment analysis were collected by approval of Twitter API. During the study, a total of 310,341 tweets were collected in three periods, November 2019, May 2020 and October 2020. Tweets are classified according to 10 different emotions. One of the main findings of the study is that “positive”, “trust” and “anticipation” feelings are at the top of the tweets shared about these institutions under normal conditions and crisis conditions. Sentiment consistency was tested with Friedman test for each institution after emotional analysis was performed in all institutions (p<0.05). In all institutions except FAO, a significant relationship was found between the emotion medians of the feeling of “fear” according to the periods. However; according to the results of the sentiment analysis, an increase was observed in the “fear” feeling of the institutions during the pandemic period. This research; non-profit international organizations that make environment, health and sustainable development a principle; provides a conceptual framework for understanding when and why follower behavior originates. Thus, it can enable organizations to realize their missions, create meaningful support in times of crisis, increase benefit and awareness, facilitate decision-making, and increase communication and interaction