Background Check

This is a framework to create and evaluate classifier models with confidence levels.

For a better explanation of the method go to the official web-page of the paper.

Supplementary material

The suplementary material can be found in icdm2016.

Evaluate classifier with confidence given two thresholds
Evaluate classifier with confidence with volume under the Precision Recall Gain and ROC curve
Select optimal threshold for given deployment specification

MIT

Name		Name	Last commit message	Last commit date
Latest commit History 275 Commits
cwc		cwc
datasets		datasets
icdm2016		icdm2016
jupyter		jupyter
utils		utils
.gitignore		.gitignore
README.md		README.md
compare_mvn.py		compare_mvn.py
contributors.txt		contributors.txt
diary.py		diary.py
digits_vs_letters.py		digits_vs_letters.py
results_Ferri2004.py		results_Ferri2004.py
results_Hempstalk2008.py		results_Hempstalk2008.py
results_Krawczyk2015.py		results_Krawczyk2015.py
results_Li2014.py		results_Li2014.py
results_Orallo2011.py		results_Orallo2011.py
results_Tax2008.py		results_Tax2008.py
results_Tax2008_multiclass.py		results_Tax2008_multiclass.py
save_hidden_representation.py		save_hidden_representation.py
summarize_results.sh		summarize_results.sh
test_chain_synthetic.py		test_chain_synthetic.py
test_chars74k.py		test_chars74k.py
test_different_mvns.py		test_different_mvns.py
test_generate_rej_given_data.py		test_generate_rej_given_data.py
test_mnist.py		test_mnist.py
test_optimality.py		test_optimality.py
test_real_datasets.py		test_real_datasets.py
test_rgrpg_synthetic.py		test_rgrpg_synthetic.py
test_volume_calculation.py		test_volume_calculation.py