Perform feature engineering

[jo-elimu]

Extract useful features from the original dataset, e.g. total word count per storybook, etc.

With each new feature, run the model training (#3) again, and see how it affects the prediction score.

• Chapter count
• Paragraph count
• Word count
• Average word length
• Average word count per chapter

---

• Depends on Split the data into a training set and test set #3

[jo-elimu]

Including avg_word_length worsened the Mean Absolute Error (MAE) from 0.3333333333333333
to 0.5833333333333334.