Updating an important but stale Python library
The project is based on a “profanity-check” library created by Victor Zhou. You can read more about it here and find it online here: https://github.com/vzhou842/profanity-check. Firstly, we installed the library in a virtual environment and experimented with different samples. We tested the model with an internal dataset consisting of 850 tweets retrieved through Twitter’s sampling API then labeled manually. This produced the following results: Confusion Matrix ActualPredicted Not Profane (0) Profane (1) Not Profane (0) 703 14 Profane (1) 93 39 Accuracy Score: 87....