Social Media Analytics: Literature Review and Directions for Future Research

Businesses are currently using social media analytics (SMA) to develop insights for improving performance and productivity across different functions. The SMA knowledge is growing diversely, and there is a need to understand the trends and approaches holistically. The present paper offers a comprehensive review of the SMA empirical literature and directions for future research. The review is based on 54 papers selected out of 843 search results. The review focuses on different domains: industrial domains, data-mining objectives, use cases, and applications. Out of the studies, public administration and consumer discretionary sectors are the dominant ones with Twitter data being used in most of the analysis. Out of the possible techniques, classification techniques and regression models are more popular than others. Stakeholder engagement is the most focused theme in the research studies. The review also offers insights into which analytical approaches are being used in which industrial domains for specific decision making. It further suggests that novel methods, such as cross-media data classification, tags detection, label priority ranking, tweeting activity signatures, and geospatial data processing have been used less and could be further explored in future research. The review also offers implications for the decision sciences domain.