{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/d178718afb414d7b99a2610a991775a9\" frameborder=\"0\" width=\"1512\" height=\"1134\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1134,"width":1512,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1134,"thumbnail_width":1512,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/d178718afb414d7b99a2610a991775a9-00001.gif","duration":295.119,"title":"CyberBuddy: Building a YouTube Spam Filtering System","description":"Hi, my name is Socket and I made a map called CyberBuddy during this hackathon. CyberBuddy is a YouTube spam filtering system that utilizes a multinomial naive base machine learning classification model. I trained the model on a Kaggle dataset of YouTube spam data and used a TF-IDF vectorizer to prepare the data for machine learning. In this video, I explain the process of creating the model, testing it with sample input, and integrating it with the YouTube API for comment deletion. Watch to learn more about the development of CyberBuddy."}