Decision Tree Regression

The algorithm used for the prediciton of the transcoding times for this project was decison tree regression. The algorithm was trained on a dataset of roughly 16000 entries collected from the University of California Irvine Machinie Learning Repository. The model has an r^2 score of 0.9813526253175591. Below are some figures depicting elements of the mode

Prediction Error

Partial Dependencies

First 3 Levels of Tree Model

Full Tree Model