Spam-Classifier

A spam classifier to predict spam and ham(not spam) Emails using concepts of Machine Learning, Natural Language Processing(NLP) and Python.

Workflow of the project

1. Loading Dataset

• Dataset - https://archive.ics.uci.edu/ml/datasets/SMS+Spam+Collection

2. Data Visualization

3. Data Preprocessing

• Data cleaning (using RegEx)
• Tokenization (using Word Tokenization)
• Removing Stop words
• Lemmatization (using WordNet)
• Vectorization (using TF-IDF)
• Label Encoding

4. Splitting Dataset into Training and Testing set

5. Model Training

• Naïve Bayes
• Random Forest
• Support Vector Machine
• k- Nearest Neighbors

6. Model Evaluation

• Cross Validation Scores
• Accuracy on Testing and Testing dataset

Accuracy reported on various Algorithms used:

References

1. Machine Learning Mastery - https://machinelearningmastery.com/natural-language-processing/
2. TF-IDF - https://towardsdatascience.com/tf-idf-for-document-ranking-from-scratch-in-python-on-real-world-dataset-796d339a4089
3. Stemming vs Lemmatization - https://nlp.stanford.edu/IR-book/html/htmledition/stemming-and-lemmatization-1.html
4. Stop Words - https://kavita-ganesan.com/what-are-stop-words/#.YWFEfNpBxPY
5. RegEx basics - https://docs.python.org/3/howto/regex.html