A Preliminary Numerical Study to Compare the Physical Method and Machine Learning Methods Applied to GPR Data for Underground Utility Network Characterization.
Rakeeb Mohamed JauferAmine IhamoutenYann GoyatShreedhar Savant TodkarDavid GuilbertAli AssafXavier DérobertPublished in: Remote. Sens. (2022)
Keyphrases
- machine learning methods
- statistical methods
- synthetic data
- data sets
- machine learning
- missing data
- correlation analysis
- database
- input data
- statistical analysis
- test data
- prior knowledge
- bayesian methods
- experimental study
- sensitivity analysis
- data collection
- network structure
- preprocessing
- receiver operating characteristic curves
- data analysis
- numerical data
- machine learning approaches
- information loss
- ensemble methods
- noisy data
- data sources
- missing values
- network model
- association rules
- pairwise
- artificial neural networks
- numerical methods
- data transfer
- training samples
- peer to peer
- learned knowledge
- probabilistic model