Combatting Toxicity
moderate and prevent verbal toxicity
Jessica Patel
Project Description
The world of multiplayer video games has reached a point where player communities have become plagued with verbal toxicity in the form of hate and harassment, a detrimental issue for both players and game development companies alike. This project involves developing an intervention system that can be embedded into video games to detect, moderate and prevent verbal toxicity. The intervention system, “Combatting Toxicity”, was built using natural language processing (NLP) machine learning techniques to investigate toxicity detection methods and design thinking as a methodology to promote a user-centered approach.
The NLP algorithms conduct back-end processing to detect and analyze two modalities of toxic speech; text transcription and audio features. The intervention system then reacts to toxicity through interface elements that are overlaid in gameplay to moderate and prevent verbal toxicity.
The presented prototype displays an iterated, tested and polished version of the intervention system. You are welcome to explore the intervention system by speaking into the microphone to test the toxicity detection, and by interacting with the various responses of the system.
BIO
Jessica is an inquisitive thinker and thrives in tackling challenging problems. Her journey began during her bachelor’s degree in physics and astronomy. While doing research on ultra-faint dwarf galaxies, she realized that she wanted to work towards something that would allow her to produce more tangible results and solutions for the complex problems that affect people everyday. That's when she pursued the dynamic intersection of UX design, game research and programming. She then achieved a certification in UX design, and is currently completing a Master of Design where she is working on a thesis project involving human-computer interaction, game studies, and machine learning.