Explore feature engineering techniques to incorporate text information as indicator variables for boosted tree models in this 52-minute screencast. Learn how to predict housing prices in Austin, TX using tidymodels and xgboost, based on data from the #SLICED semifinals. Follow along with exploratory visualization, tokenization, linear modeling, and advanced feature engineering. Dive into the process of running and tuning the model to achieve optimal results. Gain practical insights into data science workflows and machine learning applications in real estate analysis.
Overview
Syllabus
Introduction
Data
Exploratory visualization
Plot Austin
Tokenization
Top Words
Linear models
Model tidy
Model words
Model shape
Data set
Filter by scale
Modeling
Feature Engineering
Running the model
Tuning the model
Tuning the model again
Results
Taught by
Julia Silge
Related Courses
-
Introduction to Machine Learning
-
Tuning XGBoost Using Tidymodels
-
Build Features for Machine Learning from Netflix Description Text
-
Foundations of Machine Learning
-
Tune XGBoost With Early Stopping to Predict Shelter Animal Status
-
Feature Engineering & Interpretability for XGBoost with Board Game Ratings
