Contributers: Taewon Hwang, Ahreum Lee, Chaeyeon Kang, Seol Han

Purpose: Eating disorders (EDs) are a growing public health concern, significantly impacting individuals' physical, psychological, and emotional well-being. Studies show a rise in ED prevalence, particularly among younger demographics, with media playing a critical role in exacerbating these disorders. This project aims to address this issue by developing censoring tool that limits media content related to extreme diets, creating a safer online environment for vulnerable individuals, especially young audiences.

Achievements: We used different methods of preprocessing and machine learning models to test and compare results. Stop word removal along with Naive Bayes classifier proved to be the most effective. It achieved a maximum accuracy of 90% with the Naive Bayes classifier.

Description: As one of the biggest platforms for various groups to consume media, YouTube has one solution for censoring potentially harmful content: manual flagging by users. Taking one step further, the AKCSE Life Science (LS) and Computer Science (CS) divisions collaborated to develop a Chrome extension powered by machine learning (ML), specifically to automatically moderate YouTube content and help prevent ED-triggering videos.