Wrap-up video

Predicting CTR with Machine Learning in Python

Kevin Huo

Instructor

Chapter 1

  • Introduction to CTRs

    • Learned about basic problem from a classification lens

  • Overview of machine learning models

    • Practiced with logistic regression on various datasets

  • Brief overview of CTR prediction

    • Applied decision trees for CTR prediction
Predicting CTR with Machine Learning in Python

Chapter 2

  • Basic exploratory data analysis

    • Looked at specific features and variability with CTR

  • Feature engineering

    • Learned hashing and created features from existing ones

  • Standardization

    • Applied standard scaling and log normalization
Predicting CTR with Machine Learning in Python

Chapter 3

  • Applications of metric evaluation

    • Learned the business interpretations of evaluation metrics through confusion matrices and an ROI framework

  • Model evaluation

    • Evaluated precision and recall relative to a baseline classifier

  • Tuning models

    • Learned concepts of regularization and cross-validation

  • Ensembles and hyperparameter tuning

    • Tuned hyperparameters using grid search for a Random Forest
Predicting CTR with Machine Learning in Python

Chapter 4

  • Basic concepts and model

    • Learned the inner workings of neural networks

  • Hyperparameter tuning

    • Tuned hyperparameters of neural networks via hidden layers and max iterations

  • Model evaluation

    • Computed F-beta scores and implications of precision versus AUC of ROC curve

  • Model review and comparison

    • Reviewed all models covered and compared them on all evaluation metrics
Predicting CTR with Machine Learning in Python

Thank you!

Predicting CTR with Machine Learning in Python

Preparing Video For Download...