The proliferation of online gambling site promotions within YouTube comment sections has become a concerning phenomenon, posing potential harm to users through manipulative content and illegal links. This study aims to design and implement an automated detection and reporting system for online gambling spam comments using a Natural Language Processing (NLP) approach and the Random Forest algorithm. The system is developed as a browser extension with a client-server architecture to enable direct, real-time intervention. Data was collected through web scraping techniques from YouTube comment sections on trending gaming content. The preprocessing pipeline includes case folding, filtering, tokenization, stopword removal, and vectorization using TF-IDF. The results demonstrate that the Random Forest algorithm successfully classifies gambling spam comments with a confidence score reaching 95.00%. Functional testing using Blackbox Testing across 10 primary scenarios achieved a 100% success rate. The main contribution of this research is a reporting feature directly integrated with the YouTube interface, allowing users to report spam without leaving the browser. This system provides a practical solution for both users and content creators to effectively mitigate the spread of online gambling promotions.

Keywords: Spam Detection, Online Gambling, YouTube, Random Forest, Natural Language Processing, TF-IDF, Browser Extension