Back to Projects

Machine Learning

Logistic Regression Text Classification

A tweet text classification project that moves from a dummy baseline to logistic regression with bag-of-words features, hyperparameter tuning, and coefficient interpretation.

Period

May 2026

Tools / Tech

PythonScikit-learnLogistic RegressionCountVectorizerGridSearchCVCross-validation

Why I built it

The goal was to practice building a defensible baseline before tuning a text classifier.

Links

What it includes

  • Starts with a dummy classifier baseline.
  • Uses CountVectorizer and logistic regression for bag-of-words classification.
  • Tunes vectorization and regularization settings with cross-validation.

What I worked on

  • Built baseline and logistic-regression workflows.
  • Compared tuned parameters and interpreted model coefficients.
  • Completed the text-classification notebook myself and used evaluation results to judge whether changes improved the classifier.

What I Learned

  • Learned why a simple baseline matters for text classification.
  • Practiced connecting coefficients to interpretable words.
  • Improved comfort with cross-validation and hyperparameter search.