Scikit-learn was utilized to predict the likelihood of a criminal reoffending, using the COMPAS dataset. Data cleaning procedures involved class rebalancing techniques such as SMOTE and filling in null values. Various classical machine learning models were experimented with before ultimately settling on the random forest model, with a 93.5% accuracy result.
-
Notifications
You must be signed in to change notification settings - Fork 0
giorgioandronico/recidivity-binary-classification
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Binary classfication with scikit-learn and Random Forest models.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published