The objective of this project is to detect spam messages. The dataset includes tagged SMS messages. It contains 5,574 English SMS messages in total; tagged as ham or spam. Check the SMSSpamCollection dataset for details.

• MLlib with PySpark is used to build the model.
• Preprocessing steps are performed and feature engineering is applied using TF-IDF.
• Logistic regression, Random Forest and Naive Bayes are used for classifications.
• The best performance is achieved for the Logistic regression with 97% accuracy.

This Project’s GitHub Repository

• Big Data
• Spark
• MLib
• Classification