README.md

DataMiningHW2

11 models for Polish+companies+bankruptcy+data

Table 1 Final Results of all-5-years data for Comparative Models with unbalanced training data

ModelName	Accuracy	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.897	0.627	0.046	0.542	0.011
GaussianNB	0.863	0.66	0.049	0.531	0.012
SVM	0.976	0.5	0.024	0.494	1.470
DecisionTree	0.965	0.657	0.114	0.647	0.750
SGD	0.876	0.642	0.046	0.533	0.042
Nearest_Neighbors	0.723	0.675	0.042	0.467	0.007
AdaBoost	0.978	0.582	0.131	0.628	7.771
GradientBoosting	0.978	0.588	0.145	0.636	6.518
HistGradientBoosting	0.982	0.64	0.262	0.708	9.174
MLP	0.977	0.623	0.158	0.667	113.919

Table 1year.arff with unbalanced training data

ModelName	Accuracy	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.954	0.010	0.505	0.050	0.499	0.005
GaussianNB	0.739	0.594	0.670	0.078	0.508	0.005
SVM	0.954	0.000	0.500	0.046	0.488	0.400
DecisionTree	0.959	0.531	0.755	0.313	0.759	0.245
SGD	0.955	0.010	0.505	0.055	0.499	0.027
NearestCentroid	0.685	0.656	0.671	0.075	0.483	0.004
AdaBoost	0.962	0.271	0.633	0.229	0.687	2.649
GradientBoosting	0.968	0.385	0.691	0.345	0.754	2.078
HistGradientBoosting	0.974	0.490	0.744	0.466	0.811	120.715
MLP	0.953	0.312	0.648	0.178	0.675	42.656
LogisticRegression(Ours)	0.710	0.143	0.430	0.011	0.421	1.581

Table 2year.arff with unbalanced training data

ModelName	Accuracy	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.961	0.000	0.499	0.038	0.490	0.007
GaussianNB	0.827	0.243	0.547	0.043	0.500	0.007
SVM	0.962	0.000	0.500	0.038	0.490	1.047
DecisionTree	0.954	0.409	0.692	0.182	0.688	0.367
SGD	0.962	0.000	0.500	0.038	0.490	0.040
NearestCentroid	0.644	0.626	0.635	0.054	0.447	0.006
AdaBoost	0.963	0.070	0.534	0.078	0.553	3.831
GradientBoosting	0.961	0.096	0.545	0.075	0.568	3.053
HistGradientBoosting	0.978	0.435	0.717	0.431	0.792	49.879
MLP	0.950	0.261	0.619	0.109	0.629	59.709
LogisticRegression(Ours)	0.787	0.231	0.514	0.018	0.458	1.690

Table 3year.arff with unbalanced training data

ModelName	Accuracy	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.957	0.000	0.500	0.043	0.489	0.007
GaussianNB	0.831	0.336	0.594	0.059	0.525	0.015
SVM	0.957	0.000	0.500	0.043	0.489	1.276
DecisionTree	0.930	0.299	0.628	0.101	0.614	0.438
SGD	0.957	0.000	0.500	0.043	0.489	0.046
NearestCentroid	0.697	0.522	0.614	0.058	0.472	0.005
AdaBoost	0.956	0.112	0.552	0.084	0.577	4.065
GradientBoosting	0.959	0.134	0.565	0.117	0.599	3.281
HistGradientBoosting	0.968	0.313	0.655	0.292	0.720	101.454
MLP	0.948	0.224	0.602	0.108	0.620	60.222
LogisticRegression(Ours)	0.872	0.182	0.536	0.028	0.499	1.868

Table 4year.arff with unbalanced training data

ModelName	Accuracy	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.948	0.007	0.503	0.058	0.493	0.007
GaussianNB	0.864	0.464	0.675	0.112	0.593	0.013
SVM	0.948	0.000	0.500	0.052	0.487	0.956
DecisionTree	0.948	0.516	0.744	0.283	0.740	0.451
SGD	0.948	0.007	0.503	0.055	0.493	0.039
NearestCentroid	0.718	0.654	0.688	0.093	0.512	0.005
AdaBoost	0.960	0.307	0.652	0.285	0.712	3.762
GradientBoosting	0.948	0.000	0.500	0.052	0.487	2.987
HistGradientBoosting	0.969	0.444	0.721	0.437	0.792	96.038
MLP	0.946	0.340	0.660	0.195	0.684	56.011
LogisticRegression(Ours)	0.924	0.179	0.564	0.046	0.547	1.787

Table 5year.arff with unbalanced training data

ModelName	Accuracy	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.932	0.033	0.516	0.085	0.514	0.004
GaussianNB	0.884	0.554	0.731	0.200	0.665	0.004
SVM	0.941	0.157	0.578	0.193	0.617	0.347
DecisionTree	0.937	0.595	0.779	0.347	0.765	0.183
SGD	0.935	0.050	0.525	0.114	0.530	0.024
NearestCentroid	0.775	0.736	0.757	0.162	0.587	0.007
AdaBoost	0.946	0.430	0.707	0.322	0.746	2.211
GradientBoosting	0.951	0.455	0.721	0.371	0.768	1.697
HistGradientBoosting	0.962	0.537	0.765	0.485	0.818	110.560
MLP	0.933	0.537	0.750	0.307	0.744	36.765
LogisticRegression(Ours)	0.925	0.421	0.682	0.109	0.621	1.026

Table 1year.arff with balanced training data via SMOTE

ModelName	Precision	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.833	0.292	0.575	0.058	0.522	0.008
GaussianNB	0.506	0.771	0.632	0.063	0.390	0.013
SVM	0.812	0.635	0.728	0.108	0.564	5.137
DecisionTree	0.923	0.500	0.722	0.171	0.666	0.525
SGD	0.903	0.438	0.681	0.121	0.620	0.073
NearestCentroid	0.685	0.635	0.661	0.073	0.481	0.005
AdaBoost	0.903	0.542	0.731	0.154	0.643	5.270
GradientBoosting	0.905	0.542	0.732	0.156	0.646	4.317
HistGradientBoosting	0.974	0.510	0.753	0.461	0.813	25.384
MLP	0.900	0.562	0.740	0.157	0.643	80.322
LogisticRegression(Ours)	0.668	0.286	0.479	0.012	0.410	20.871

Table 2year.arff with balanced training data via SMOTE

ModelName	Precision	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.780	0.365	0.581	0.048	0.493	0.009
GaussianNB	0.595	0.626	0.610	0.050	0.421	0.019
SVM	0.758	0.609	0.686	0.070	0.509	11.095
DecisionTree	0.910	0.435	0.682	0.105	0.609	0.694
SGD	0.796	0.487	0.647	0.063	0.518	0.118
NearestCentroid	0.640	0.617	0.629	0.053	0.444	0.007
AdaBoost	0.862	0.470	0.673	0.081	0.564	7.616
GradientBoosting	0.865	0.496	0.687	0.088	0.571	6.333
HistGradientBoosting	0.973	0.400	0.698	0.334	0.757	31.196
MLP	0.885	0.522	0.710	0.106	0.596	104.437
LogisticRegression(Ours)	0.643	0.692	0.667	0.029	0.422	19.106

Table 3year.arff with balanced training data via SMOTE

ModelName	Precision	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.777	0.388	0.591	0.056	0.500	0.011
GaussianNB	0.688	0.590	0.641	0.064	0.474	0.023
SVM	0.778	0.567	0.677	0.079	0.525	11.070
DecisionTree	0.892	0.403	0.659	0.095	0.592	0.804
SGD	0.890	0.455	0.682	0.106	0.600	0.113
NearestCentroid	0.694	0.522	0.612	0.058	0.471	0.004
AdaBoost	0.849	0.470	0.668	0.086	0.563	8.036
GradientBoosting	0.859	0.440	0.659	0.084	0.566	6.780
HistGradientBoosting	0.963	0.410	0.699	0.268	0.733	24.232
MLP	0.879	0.388	0.644	0.084	0.574	106.323
LogisticRegression(Ours)	0.650	0.682	0.666	0.040	0.436	19.910

Table 4year.arff with balanced training data via SMOTE

ModelName	Precision	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.795	0.477	0.645	0.086	0.539	0.010
GaussianNB	0.802	0.588	0.701	0.108	0.561	0.022
SVM	0.801	0.641	0.725	0.119	0.568	9.120
DecisionTree	0.889	0.484	0.698	0.139	0.626	0.726
SGD	0.761	0.758	0.760	0.125	0.553	0.101
NearestCentroid	0.719	0.654	0.688	0.093	0.512	0.006
AdaBoost	0.849	0.608	0.735	0.139	0.606	7.405
GradientBoosting	0.862	0.647	0.761	0.161	0.626	6.209
HistGradientBoosting	0.965	0.569	0.778	0.421	0.805	30.426
MLP	0.878	0.523	0.710	0.139	0.621	102.660
LogisticRegression(Ours)	0.754	0.643	0.700	0.064	0.501	22.184

Table 5year.arff with balanced training data via SMOTE

ModelName	Precision	Recall	ROC_AUC	PR_AUC	F1_score	Time_Used
BernoulliNB	0.799	0.620	0.716	0.147	0.590	0.013
GaussianNB	0.873	0.678	0.782	0.229	0.674	0.007
SVM	0.867	0.711	0.795	0.233	0.674	2.447
DecisionTree	0.896	0.554	0.738	0.219	0.682	0.426
SGD	0.892	0.694	0.800	0.265	0.703	0.053
NearestCentroid	0.773	0.736	0.755	0.160	0.585	0.002
AdaBoost	0.905	0.702	0.811	0.294	0.724	4.232
GradientBoosting	0.901	0.678	0.797	0.276	0.714	3.388
HistGradientBoosting	0.953	0.636	0.806	0.444	0.811	25.816
MLP	0.911	0.587	0.761	0.261	0.712	61.952
LogisticRegression(Ours)	0.813	0.789	0.802	0.112	0.561	16.805