Dataset research report
Compas research report
A reproducible data report with schema notes, generated chart evidence, suggested follow-up questions, and export-ready Helix queries.
Executive Summary
Compas The Compas dataset for recidivism prediction. Dataset known to have racial bias issues, check this Propublica article on the topic. Configurations and tasks Configuration Task Description encoding Encoding dictionary showing original values of encoded features. two-years-recidividity Binary classification Will the defendant be a violent recidivist? two-years-recidividity-no-race Binary classification As above, but the race feature is… See the full description on the dataset page: https://huggingface.co/datasets/mstz/compas.
Research Context
Compas: 500 rows by 14 columns. These exploratory charts are generated automatically from the data - open the dataset in Helix to ask your own questions.
Data Profile
Chart Evidence
These views are generated from the dataset profile. Each chart is paired with a Helix query so it can be opened, adjusted, and exported.
age vs number_of_juvenile_fellonies
age vs number_of_juvenile_fellonies, coloured by is_male.
Open and export this chartCorrelation of numeric columns
Pearson correlation between numeric columns.
Open and export this chartFollow-Up Queries
Preview Rows
| # | is_maleboolean | ageinteger | racetext | number_of_juvenile_felloniesinteger | decile_scoreinteger | number_of_juvenile_misdemeanorsinteger | number_of_other_juvenile_offensesinteger | number_of_prior_offensesinteger |
|---|---|---|---|---|---|---|---|---|
| 1 | True | 69 | Other | 0 | 1 | 0 | 0 | 0 |
| 2 | True | 34 | African-American | 0 | 3 | 0 | 0 | 0 |
| 3 | True | 44 | Other | 0 | 1 | 0 | 0 | 0 |
| 4 | True | 43 | Other | 0 | 4 | 0 | 0 | 3 |
| 5 | False | 39 | Caucasian | 0 | 1 | 0 | 0 | 0 |
| 6 | True | 21 | Caucasian | 0 | 3 | 0 | 0 | 1 |
Data Dictionary
- is_male bool
- age numeric
- race categorical
- number_of_juvenile_fellonies numeric
- decile_score numeric
- number_of_juvenile_misdemeanors numeric
- number_of_other_juvenile_offenses numeric
- number_of_prior_offenses numeric
- days_before_screening_arrest datetime
- is_recidivous bool
- days_in_custody datetime
- is_violent_recidivous bool
- violence_decile_score numeric
- two_year_recidivous datetime
Method And Limits
- Load the catalog entry and preview rows from the processed dataset file.
- Infer numeric, categorical, time, and location fields from real columns.
- Generate a small set of defensive Plotly chart specifications from that profile.
- Expose each chart idea as a query link so the report can be rerun or exported in Helix.
This report is intentionally reproducible. It uses the local catalog metadata and generated chart specifications rather than claiming external conclusions beyond the dataset.