Automatic content analysis of social media short texts: scoping review of methods and tools

Content analysis is a widely applied method, applicable to qualitative and quantitative data. In content analysis, computer programs could be used not only for manual typing of codes and categories but for automatic screening of texts, software-assisted identifying and coding of words, phrases, paragraphs or events. Methods for analyzing social media textual data are usually associated with computer science, social media and communication scholar articles. These empirical sources focus on optimizing the goals of computer sciences and mostly evaluate the percentage of documents, phrases or words correctly categorized into a set of themes. Our study aimed to scope the nature and extent of empirical research articles and online materials around automatic content analysis not limited to computer science, media, and communication. This article analyzed and systematized the empirical articles of the last five years examining the categories of automatic content analysis and the related research questions focusing on social sciences and provided practical examples of methods, tools and empirical applications. Research gaps were identified and recommendations for the application of automated content analysis are providedBaltijos pažangių technologijų institutasEdukologijos tyrimų institutasInformatikos fakultetasTaikomosios informatikos katedraVytauto Didžiojo universiteta