Classification of toxic comments in video games

• Autores: Luque Nieto, Luis Fernando, Portugal Carpio, Elmerson Ramith
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad La Salle
• Repositorio: Revistas - Universidad La Salle
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.ulasalle.edu.pe:article/124
• Enlace del recurso: https://revistas.ulasalle.edu.pe/innosoft/article/view/124; https://doi.org/10.48168/innosoft.s15.a124; https://purl.org/42411/s15/a124; https://n2t.net/ark:/42411/s15/a124
• Nivel de acceso: acceso abierto
• Materia: Machine learning; Toxic chat; Natural language processing; Video games; Aprendizaje automático; Chat tóxico; Procesamiento de lenguaje natural; Videojuegos

Toxicity can have a major impact on player engagement and satisfaction. It is a complex phenomenon that has diverse causes and consequences. Among the most common causes are anonymity, competitiveness and frustration. The consequences can be serious, such as harassment, abandonment of the game and psychological damage. Gaming companies are working to find ways to address forms of toxicity on their platforms. One of the most common interactions with toxicity occurs in chat windows or in-game messaging systems. The proposed work is to pull some chat messages that occur in these "lobbies" or take them offline so that they can be categorized to determine if the player who wrote in the chat committed an infraction and depending on the category take action on the case.