*Article* **Topic Modeling and User Network Analysis on Twitter during World Lupus Awareness Day**

**Salvatore Pirri 1,\*, Valentina Lorenzoni 1, Gianni Andreozzi 1, Marta Mosca 2 and Giuseppe Turchetti 1**


Received: 20 June 2020; Accepted: 24 July 2020; Published: 28 July 2020

**Abstract:** Twitter is increasingly used by individuals and organizations to broadcast their feelings and practices, providing access to samples of spontaneously expressed opinions on all sorts of themes. Social media offers an additional source of data to unlock information supporting new insights disclosures, particularly for public health purposes. Systemic lupus erythematosus (SLE) is a complex, systemic autoimmune disease that remains a major challenge in therapeutic diagnostic and treatment management. When supporting patients with such a complex disease, sharing information through social media can play an important role in creating better healthcare services. This study explores the nature of topics posted by users and organizations on Twitter during world Lupus day to extract latent topics that occur in tweet texts and to identify what information is most commonly discussed among users. We identified online influencers and opinion leaders who discussed different topics. During this analysis, we found two different types of influencers that employed different narratives about the communities they belong to. Therefore, this study identifies hidden information for healthcare decision-makers and provides a detailed model of the implications for healthcare organizations to detect, understand, and define hidden content behind large collections of text.

**Keywords:** social media; Twitter; systemic lupus erythematosus (SLE); network analysis; topic modeling; text analysis
