Climate Change and Food Security: The Role of Natural Language Processing and Machine Learning in Automating the Analysis of Social Media Data to Discern Public Opinion of Agriculture’s Impact on the Environment

The future of agricultural production is inextricably linked to policies that are designed to mitigate climate change and the impact of agriculture on the environment. While these policies are crafted with the long-term goal of ensuring a sustainable food supply, there is often concern that these policies may threaten food security in the interim. Conversely, proponents of more progressive climate change mitigative measures have criticized current policies as “too little, too late.” Understanding public sentiment of climate change as it relates to agricultural production is imperative to developing effective policies that address climate change concerns while also ensuring a stable food supply within the current realities of agricultural production and food security needs. In response to the need to better understand public sentiment surrounding climate change in general, Reddit released the Reddit Climate Change Dataset. Within this dataset, 232,575 comments reference agriculture or food security and their relationship with climate change directly. To analyze this large dataset, this study presents the use of natural language processing (NLP) and machine learning (ML) to identify core themes that emerge in social media conversations surrounding climate change, agriculture, and food security, and how these themes change over time and across stakeholder groups.