Analysis of Sentiments in Corporate Twitter Communication – A Case Study on an Issue of Toyota

Abstract

Knowing about communication of specific issues in social media has become increasingly important for the reactive and proactive stakeholder-communication of enterprises. Tools have been designed to monitor social media sites and to aggregate data of discussions in social media. However, these tools do not consider the dynamics of discussions and are not able to reflect sentiments within these discussions. In our contribution, we address these aspects by analyzing a data set of around 730,000 Tweets published in a time frame of 19 weeks. Within this data set, we analyzed those Tweets dealing with the corporate crisis of Toyota in 2010. We classified sentiments by using a linguistic approach. In this context, we identified and investigated specific stages of communication (“quiet stages” and “peaks”). Additionally, our study concentrates on the sentiments found in Tweets of the ten most active participants of the discussion.