Skip to content

Commit

Permalink
Merge pull request #23 from srlearn/boston-housing-dep
Browse files Browse the repository at this point in the history
🗑️ Deprecate `boston_housing`
  • Loading branch information
hayesall authored Nov 2, 2022
2 parents e1c9296 + 8353ded commit 952ca15
Showing 1 changed file with 12 additions and 0 deletions.
12 changes: 12 additions & 0 deletions srlearn/boston_housing/README.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,17 @@
# boston_housing

!!! warning end
The Boston Housing dataset is deprecated. It is included here for backwards compatibility and reproducing results in old publications, but should not be used for benchmarking future results.

The dataset contains a variable `B` which is ethically problematic. The original dataset authors assumed that Black neighbors were undesirable, and that this would affect housing prices. However, this assumption was encoded in a way that makes it impossible to analyze further.

We recommend the "California Housing" dataset instead.

**See also**:

- M Carlisle, "[racist data destruction?](https://medium.com/@docintangible/racist-data-destruction-113e3eff54a8)" Medium.com, retrieved: 2022-11-02
- [sklearn.datasets.load_boston (archived)](https://web.archive.org/web/20221014215704/https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_boston.html)

"Boston Housing" is a common benchmark dataset for regression.

## Task
Expand Down

0 comments on commit 952ca15

Please sign in to comment.