•Cleaned and analysed a dataset of 31000 tweets containing normal and sexist/racial tweets.
•Used BOWandTFIDF features for word vectorization and Word2Vec and Doc2Vec for word embeddings.
•Applied different classification algorithms like logistic regression, support vector machines,random forest and XGBoost algorithm with each of the word embeddings.
•Analysed the result for each combination with the help of F1-score and performed hyperparameter tuning using gridsearch cross validation to improve the mode