Dynamic model updating (DMU) approach for statistical learning model building with missing data

BMC Bioinformatics

Table 4 Performance of different methods on the real datasets

Scenario	Dataset	MSE (Method)/MSE (Mean Imputation)
Scenario	Dataset	SLR	kNN	SLRM	RF	DMU
1	I	0.36	1.05	0.37	2.41	0.16
2	I	0.34	1.08	0.43	1.56	0.15
3	I	0.43	0.95	0.74	1.11	0.07
4	I	–	–	1.69	1.42	0.71
5	I	–	–	0.91	1.27	0.19
6	I	–	–	1.18	1.51	0.05
7	II	0.84	0.96	0.84	1.00	0.84
8	II	0.33	0.99	0.38	0.62	0.32
9	II	0.25	0.88	0.31	0.57	0.24
10	II	–	–	0.44	0.85	0.35
11	II	–	–	0.87	0.58	0.33
12	II	–	–	0.69	0.50	0.36

SLR Simple Linear Regression, KNN k Nearest Neighbors based Imputation, SLRM Simple Linear Regression combined with imputation, RF Random Forest-based Imputation, DMU Dynamic Model Updating

ISSN: 1471-2105