Big data visualization through the lens of Peirce’s visual sign theory

Data from social media platforms, such as Twitter and Facebook, are generated by people who produce, spread, share, or exchange multimedia content. Such content may include text, images, sounds, or videos. To derive insight into the behavior of social media users, researchers often use open-source technologies to visualize data and generate models for data analytics. One of the most popular open-source applications for managing and analyzing social media data is the open-source R programming language. Friedman and Feichtinger (2017) created an R package termed ‘Peirce’s sign theory R package’ to analyze data using Peirce’s principles of discovery. Though Peirce semiotics have been introduced in the context of computer programming languages, so far, no previous work has applied Peirce’s sign theory to data modelling of social media data. In this paper, we use Peirce’s sign theory R package as an overall framework to gain insight into data collected from Twitter. We assembled the data using Twitter’s Analytics algorithm, examined the relationships between variables, and visualized the results. Subsequently, we assessed the feasibility of analyzing those graphics using the triadic model set out by Jappy (2013) and Peirtarinen (2012) for the interpretation of visual signs. The study results showed that Peirce’s sign theory R package effectively analyzes and visualizes Big Data from social media feeds. However, due to complexities in both the social media data feeds and Peirce’s interpretation of meaning, as outlined by Jappy (2013) and Peirtarinen (2012), we were unable to develop algorithms that generate or suggest an interpretation of visual signs.