Development of a neural network based imputation system
MIDENET ; FESSANT
Type de document
COMMUNICATION AVEC ACTES INTERNATIONAL (ACTI)
Langue
anglais
Auteur
MIDENET ; FESSANT
Résumé / Abstract
The second part of the work-package 5 concerns the investigation of another kind of methods for correction and imputation problems based on neural networks models. Neural network models are commonly and successfully used as data analysis tool, especially for uncertain or noised data, for high dimensionality data and for non linear analysis. Our task in this project is to evaluate the benefits of using such models for imputation of missing or erroneous data in surveys. We are developing two particular neural network models and testing them for an imputation task.
