Twitter data analysis to assess the interest of citizens on the impact of marine plastic pollution

Abstract

Few studies have mined social media platforms to assess environmental concerns. In this study, Twitter was scraped to obtain a ~140,000 tweet dataset related specifically to marine plastic pollution. The goal is to understand what kind of users profiles are tweeting and how and when they do it. In addition, topic modelling and graph theory techniques have allowed us to identify main concerns on this topic: i) impact on wildlife, ii) microplastics/water pollution, iii) estimates/reports, iv) legislation/protection, and v) recycling/cleaning initiatives. Results reveal a scarce influence of organizations involved in research and marine environmental awareness, so some guidelines are depicted that could help to adjust their communication plans. This is relevant to engage society through reliable information, change habits and reinforce sustainable behaviour. A visualization tool has been created to analyze the results over time.