GitHub - rcofrer/MLWR: Malware prediction study on Microsof Malware Prediction dataset published at Kaggle.

Malware prediction project for ML course 2020 winter semester. University of Wrocław.

Feature selection:

Forward selection algorithm based on conditional mutual information as shown in Novovicova 2007 (see file in repo).

Name		Name	Last commit message	Last commit date
Latest commit History 64 Commits
Final report		Final report
Feature_selection.ipynb		Feature_selection.ipynb
Initial_EDA.ipynb		Initial_EDA.ipynb
Initial_EDA_Colab.ipynb		Initial_EDA_Colab.ipynb
LGBM_model_comparison.ipynb		LGBM_model_comparison.ipynb
LICENSE		LICENSE
NumericalVariablesPCA.ipynb		NumericalVariablesPCA.ipynb
README.md		README.md
dnn.ipynb		dnn.ipynb
dnn_feature_ranking.ipynb		dnn_feature_ranking.ipynb
entropies.pickle		entropies.pickle
featureMIs.pickle		featureMIs.pickle
initial_log_reg_model.ipynb		initial_log_reg_model.ipynb
initial_nn_model.ipynb		initial_nn_model.ipynb
kaggle_preprocessing.ipynb		kaggle_preprocessing.ipynb
kaggleprocc_lgbm_test_no_touchy.py		kaggleprocc_lgbm_test_no_touchy.py
lgbm_data_preprocessing.ipynb		lgbm_data_preprocessing.ipynb
lgbm_experiments.ipynb		lgbm_experiments.ipynb
mutualInfos.pickle		mutualInfos.pickle
nn_hyperparam_tuning.ipynb		nn_hyperparam_tuning.ipynb
proper_integer_encoding.py		proper_integer_encoding.py
repeated_baseline_lgbm.py		repeated_baseline_lgbm.py

Provide feedback