Week 6: Text classification

DATA 441/641 · Week 6

Notes

• 6.1 Text Classification Pipeline(slides)
• 6.2 Traditional Text Classification(slides)

Labs

• Lab 6: Traditional Text Classification(PDF)
• Lab 6 solutions (password)

Additional

• SVM as a QP(PDF)

Additional readings

• Practical NLP, Ch. 4 (text classification)
• sklearn: Working with text data

Discussion

• Where does leakage creep into a text classification split?
• When is logistic regression enough, and when do you want a neural net?