Missing Values and Optimal Selection of an Imputation Method and Classification Algorithm to Improve the Accuracy of Ubiquitous Computing Applications

Table 6

Factors influencing accuracy (RMSE) for each algorithm (standard beta coefficient): mean imputation.

Data characteristic	trees.J48	BayesNet	SMO	Regression	Logistic	IBk
N_attributes	−.076^**	−.075^**	−.178^**	−.072^**	.115^**	.007
N_cases	−.079^**	−.049^**	.012	−.017	−.032	−.048^**
C_imbalance	.117^**	.239^**	.264^**	.525^**	.163^**	.198^**
R_missing	.051^*	.078^**	.040	.080^**	.076^**	.068^**
SE_HS	.249^**	.285^**	.186^**	.277^**	.335^**	.245^**
SE_VS	−.009	−.013	−.006	−.013	−.016	−.010
Spread	−.382^**	−.430^**	−.261^**	−.436^**	−.452^**	−.363^**
P_missing_dum1	−.049	−.038	−.038	−.037	−.045	−.038
P_missing_dum2	−.002	.014	.002	.011	.001	.011

Note  1: N_attributes: number of attributes, N_cases: number of cases, C_imbalance: degree of class imbalance, R_missing: missing data ratio, SE_HS: horizontal scatteredness, SE_VS: vertical scatteredness, spread: missing data spread, and missing patterns: univariate (P_missing_dum1 = 1, P_missing_dum2 = 0), monotone (P_missing_dum1 = 0, P_missing_dum2 = 1), and arbitrary (P_missing_dum1 = 1, P_missing_dum2 = 1)
Note  2: RMSE indicates error; therefore, lower values are better.
Note  3: ^* < 0.05, ^** < 0.01.