I am learning horse-colic dataset.

there are 2 types of values in column 2, 1 and 9.

dose that mean a record (a horse) is 1 year old or 9 years old?


According to the documentation on the link you provided:

2: Age
1 = Adult horse
2 = Young (< 6 months)

Then 1 indicates an adult horse, and since there are no 2s in the second column of the data or test files, I'd guess that 9 indicates a horse younger than 6 months. Let's check.

Column 2 contains only the values 1 and 9. In the data file, 24 of the 300 records have a value of 9 (8%), and in the test file, 4 of 68 records have a value of 9 (6%). Similar percentages is a good sign that 9 means a young horse (instead of 2).

The missing data, of which they claim there is 30%, is marked by a ? in other columns.

To get confirmation, feel free to contact the authors. And also consider reporting the bug to UCI.

  • heavily edited after checking data
    – philshem
    May 13 '19 at 6:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.