Source of Dataset: http://spamassassin.apache.org/old/publiccorpus
• Built a classification model to classify an email as either spam or ham using the Naive Bayes algorithm
• Used Beautiful Soup, re library and email parser to extract plain text from an email and performed stemming
• Implemented a CountVectorizer and TfidfTransformer pipeline from scratch to transform emails into a sparse matrix of TF-IDF features
• Evaluated the model using the Cross-Validation technique and achieved an accuracy of 97.84% and a recall score of 91.3%
• Tools used: Scikit-learn, email parser, Beautiful Soup, re, nltk, scipy, Jupyter-Notebook