ML 2.1Lesson outlineLesson 1 of 2 in Module 2
Features Against the Target
Use per-class distributions and pairwise views to decide which features help predict the target, and which repeat each other.
Video in production.
Lesson outline: Features Against the Target
In this lesson9 parts
- 01Thirty measurements and one diagnosis
- 02The breast cancer diagnosis table
- 03Comparing class averages
- 04One feature, two distributions
- 05Reading a threshold off the CDFs
- 06Two features at once
- 07Features that repeat each other
- 08The code
- 09Recap, your challenge, and what comes next
Builds on
This lesson stands on its own. These go deeper into what it uses.
Key terms: Features Against the Target
The words this lesson introduces, each in one line. The module’s glossary collects them all, and the AI glossary has every term across the courses.
- histogram, density (reminder, Stats 1.1)
- Counts of values in bins; with density on the vertical axis, each class's bars have the same total area, so a larger class does not look more important.
- class-conditional distribution
- The distribution of a feature within one class:
concave_points_worstamong the malignant tumors. - overlap
- The region where two class distributions share values.
- signal
- Information a feature carries about the label; the less the class distributions overlap, the more signal.
- empirical CDF (reminder, Stats 1.2)
- For each value, the fraction of a group at or below it.
- CDF gap (narration)
- The largest vertical distance between the two classes' empirical CDFs: 0.812 for
concave_points_worst. A threshold placed there separates the classes as well as the feature can. - Pearson correlation (reminder, Stats 1.6 and 1.9)
- A number from −1 to 1 measuring how closely two features follow a straight line together.
- pair plot, heatmap (reminder, Stats 1.10)
- A scatter for every pair of features; a grid of cells coloured by each pair's correlation.
- redundant feature
- A feature whose information another feature already carries:
perimeter_meanbesideradius_mean. - summary statistic
- One number computed from a whole data set, such as a correlation; it can not show the shape behind it.
- skewed (narration)
- A distribution with a long tail on one side, whose few large values pull its mean away from its median:
area_se, median 24.61, largest 542.2.
Quiz: Features Against the Target
Your first pick on each question is the one that counts. Getting one wrong here is how the lesson sticks.