« Pharmakon » pour les usagers : Analyse des sentiments du récit dans WhatsApp

Although Derrida's concept of "pharmakon," which denotes the dual valence-positive and negative in oral and written texts, has been extensively investigated in literary, philosophical, and cultural studies, its application to conversations on social media platforms like WhatsApp, particularly in the context of sentiment analysis, remains underexplored. This study scrutinises a dataset of 5,727 messages exchanged by lecturers and administrative staff in the WhatsApp group of the Department of Modern Languages and Translation Studies at the University of Calabar, collected between April 30, 2019, and March 27, 2024. Using natural language processing (NLP) techniques with R, a large language model (LLM), and Microsoft Excel as digital tools, we classify both textual sentiments and participants' sentiments. The analysis aims to foster sustainable peaceful interactions, thereby enhancing productive collaboration within the group.