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

Table 9

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

Data characteristic	trees.J48	BayesNet	SMO	Regression	Logistic	IBk
N_attributes	−.080^**	−.073^**	−.176^**	−.071^**	.115^**	.007
N_cases	−.081^**	−.049^**	.012	−.018	−.034^*	−.047^**
C_imbalance	.135^**	.237^**	.261^**	.524^**	.133^**	.211^**
R_missing	.062^**	.083^**	.044	.084^**	.075^**	.070^**
SE_HS	.225^**	.275^**	.183^**	.271^**	.313^**	.254^**
SE_VS	−.009	−.013	−.006	−.013	−.014	−.010
Spread	−.365^**	−.428^**	−.265^**	−.427^**	−.441^**	−.361^**
P_missing_dum1	−.035	−.037	−.034	−.033	−.048	−.038
P_missing_dum2	.012	.015	.004	.012	−.004	.009

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.