What is this software?
This software helps in identifying hateful tweets. It can be used to report cyber-bullying.
This software currently identifies sexist and racist tweets.
How to use ALD ?
It trains itself by reading a file containing a tweet label and the tweet.
On the basis of that trained model then it predicts the label for a tweet
Note: This test file also contains the label which is used to find the accuracy of the model
Current Accuracy of Model: 0.857 . Precision = 0.54 Recall = 0.67 . on test datasets
The above results can be verified by running evaluator.py
lang-model.py is a rule based language model based on general trends
lang_model_unigram is a unigram language model
lang-model-bigram is a bigram language model
Requirements : Python3
Run Command : python3 lang-model-bigram.py