This dataset used in this project is from: https://www.kaggle.com/datasets/danofer/compass?select=cox-violent-parsed_filt.csv
This file contains preprocessing, Random Forest, SHAP and Anchors.
The following subsections contain names and links to the files that correlate to them.
There are a lot of files from exploring the data and working with the different models and methods, but the main file is:
- EDA.ipynb
- [data analysis.ipynb](data analysis.ipynb)
The features that we will include are:
- is_recid (target)
- age
- sex
- race
- juv_fel_count
- juv_misd_count
- juv_other_count
- c_charge_degree
- priors_count
- Lara Oriol Cabrera
- Jonas Dauksa
- Eduard Levinschi
- Christos Manaras
- Sarah Swinnen