A knowledge-based parser: Neural-network-based approaches. Development of a neural network based imputation system for travel diary data


Type de document
Résumé / Abstract
The deliverable is dedicated to data validation (erroneous data detection) and data correction (imputation methods) in the field of surveys. We describe experiments conducted in the scope of the MEST-TEST project for studying new statistical methods based on neural networks and their potential to provide efficient data correction tools. We show more precisely that the self organizing map can be used sucessfully for these tasks. A self organizing map is designed and calibrated according to available observations, described through a set of correlated variables handle together. The map can then be used both to detect erroneous data and to impute values to partial operations. We have show that this model gives satisfying results compared to other classical methods. Moreover, we established a state of the art of the application of non conventional neural networks on imputation problem and we experimented some original model on a concrete and real size transport survey database. We worked on the trip description database coming from the 1993-1994 French national personal travel survey. We also give results concerning the vehicle description files coming from the MEST pilot surveys., RAPPORT DE CONTRAT

puce  Accès à la notice sur le portail documentaire de l'IFSTTAR

  Liste complète des notices publiques de l'IFSTTAR