Created a pipeline of Natural language processing for finding a SMS will be a Ham or Spam.
- First the message is processed using Processing text function ie, removed any (comma, question mark etc.)
- Used Tfidf Transformer (TF-IDF stands for term frequency-inverse document frequency)
- Using MultinomialNB from Naive Bayes Classifier.
- Creating the Pipeline
pipeline = Pipeline([('Text',CountVectorizer(analyzer = processing_text)), ('tfidf',TfidfTransformer()), ('classifying', MultinomialNB())])
**If you enjoy this project, please consider to continue developing and maintaining it.**