Using meta-learning for automated algorithms selection and configuration: an experimental framework for industrial big data

Journal of Big Data

Table 7 Performance of AMLBID and the baseline tools on the benchmark datasets

Dataset	AMLBID	TPOT	Auto-sklearn(V)	Auto-sklearn(E)	Original paper result
[44]	0.9374	0.9120	0.8215	0.9283	0.85
[45]	0.9706	0.9517	0.9632	0.9356	0.95
[46]	0.9941	0.9907	0.9782	0.99	0.9895
[47]	0.9205	0.9991	0.9357	0.6863	0.9984
[48]	0.8971	0.6711	0.908	0.9723	0.9677
[49]	0.9706	0.7767	0.678	0.9843	0.9278
[50]	0.8967	0.8899	0.6783	0.7952	0.884
[51]	0.8748	0.7826	0.6702	0.7727	0.8659
Wafer-ds	0.8571	0.7312	0.8033	0.8953	–
HTRU	0.8880	0.8415	0.9027	0.6591	–
Cnae-9	0.9671	0.8803	0.7922	0.8365	–
Gas_Sens	0.9739	0.9843	0.9256	0.9468	–
Covertype	0.8344	0.7307	0.7890	0.6521	–
Kc1	0.8793	0.7097	0.7697	0.8552	–
jannis	0.6719	0.7229	0.6171	0.6845	–
MiniBooNE	0.9645	0.9423	0.8343	0.8903	–
KDDCup	0.9740	0.8934	0.9331	0.95	-
segment	0.9735	0.9681	0.9337	0.9542	–
Higgs	0.713	0.726	0.7135	0.729	–
Credi-g	0.7921	0.7188	0.5739	0.6121	–
shuttle	0.9649	0.9905	0.8429	0.9362	–
APS Failure	0.9910	0.9933	0.9716	0.984	–
nomao	0.9708	0.9570	0.6995	0.7987	–
CustSat	85.59	0.8276	0.8072	0.8290	–
kr-vs-kp	0.9976	0.9209	0.6532	0.7593	–
car	0.9754	0.9999	0.8549	0.9462	–
albert	0.8759	0.8005	0.8288	0.7981	–
airlines	0.6982	0.6758	0.7094	0.5927	–
Numerai28.6	0.5207	0.4229	0.4836	0.4433	–