Generative Adversarial Network-based Missing Data Handling and Remaining Useful Life Estimation for Smart Train Control and Monitoring Systems

Table 4

Various methods for handling missing values.

Methods for handling missing values	Detailed methods	Related research studies
Removals of data sets with missing values	(i) Ignorance of records with missing values; (ii) Data without missing values are used only for an input vector	A number of research studies including [25]
Estimation of missing values (I)	(iii) Estimation of missing values using mean, MCMC, and nearest neighbours; (iv) Estimation considering only the attribute that has missing values	Moldovan et al. [26]
Estimation of missing values (II), multiple imputation	(v) Data estimation considering overall attributes’ dependency; (vi) Missing values estimation using regression and other statistical methods	Hruschka et al. [27]; Yuan [28]
Generation of a new data set	(vii) Generative adversarial network- (GAN-) based data generation; (viii) Replacement of the data having missing values with newly generated data	Kim and Lee [23, 24]; Douzas and Bacao [29]