ContentSense

With the exponential growth of digital communication, platforms face increasing challenges in moderating harmful, offensive, or inappropriate content at scale. ContentSense is a real-time content moderation system designed to detect and flag unsafe text across websites. Built as a Chrome extension, it leverages the Google Extension API to scan front-end content directly within the user’s browser. Upon detecting text input, the extension sends it to a neural network-based model hosted on a backend server, which performs fast and accurate analysis to classify the content as either appropriate or unsafe. The system then provides immediate feedback to the user, promoting safer online interactions. This project also served as a hands-on experience in deploying a full-stack AI application, combining Google Extension API for the extension interface, FastAPI for the backend, and deep learning techniques for content classification.

The project can be accessed at:

---
https://github.com/chengq220/ContentSense
---