Kernel classification rules from missing data
From MaRDI portal
The author studied nonparametric kernel classification rules when data are missing from training sets. Two treatments of missing data are investigated: 1. Ignore the missing data and use the regular kernel classification rule on the known data. 2. Estimate the missing data by a weighted sum of known data and replace the missing data by the estimated ones and then use the regular classification rule. Convergence of missclassification probabilities is studied. Simulation and comparison studies are carried out in this paper.
Recommendations
Cited in
(13)- Classification with incomplete functional covariates
- Bandwidth choice for nonparametric classification
- Kernel classification with missing data and the choice of smoothing parameters
- On classification with nonignorable missing data
- Conditional probability estimation based classification with class label missing at random
- On nonparametric classification with missing covariates
- Handling missing values in support vector machine classifiers
- Some results on classifier selection with missing covariates
- A simulation study of imputation techniques in linear quadratic and kernel discriminant analyses
- Strong consistency of a kernel-based rule for spatially dependent data
- Sensitivity of missing values in classification tree for large sample
- A kernel PLS based classification method with missing data handling
- Method to Design Pattern Classification Model with Block Missing Training Data
This page was built for publication: Kernel classification rules from missing data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4277156)