Data mining project to predict late payment using decision tree algorithms. The problem was modeled in three steps:
- Classification of payment of invoices between: on time and late;
- Classification of late payment of invoices between: in the due month and later;
- Estimated number of days overdue for overdue invoices.
pip install -r requirements.tx
- sklearn.ensemble.AdaBoostClassifier
- sklearn.ensemble.BaggingClassifier
- imblearn.ensemble.BalancedBaggingClassifier
- sklearn.ensemble.GradientBoostingClassifier
- imblearn.ensemble.BalancedRandomForestClassifier
- sklearn.ensemble.RandomForestClassifier
- imblearn.ensemble.RUSBoostClassifier
- xgboost.XGBClassifier
- sklearn.ensemble.AdaBoostRegressor
- sklearn.ensemble.BaggingRegressor
- sklearn.ensemble.GradientBoostingRegressor
- sklearn.ensemble.RandomForestRegressor
- xgboost.XGBRegressor
python model.py --step=<STEP_NUMPER> --action=selection --estimator=<ESTIMATOR_MODULE>
python model.py --step=<STEP_NUMPER> --action=tuning --estimator=<ESTIMATOR_MODULE>
python model.py --step=<STEP_NUMPER> --action=train --estimator=<ESTIMATOR_MODULE>
python model.py --step=<STEP_NUMPER> --action=test --estimator=<ESTIMATOR_MODULE>
If this project helped in any way in your research work, feel free to cite the following paper.
Predição de Pagamentos Atrasados Através de Algoritmos Baseados em Árvore de Decisão (here)
@article{10.25286/repa.v6i5.1746,
author = {Neto, Arthur F. S. and Silva, José F. G. da and Oliveira, Glauber N. de},
title = {Predição de Pagamentos Atrasados Através de Algoritmos Baseados em Árvore de Decisão},
journal = {Revista de Engenharia e Pesquisa Aplicada (REPA)},
pages = {1-10},
month = {11},
year = {2021},
volume = {6},
number = {5},
url = {https://doi.org/10.25286/repa.v6i5.1746},
doi = {10.25286/repa.v6i5.1746},
}