Relating the Partial Dependence Plot and Permutation Feature Importance to the Data Generating Process.
Christoph MolnarTimo FreieslebenGunnar KönigGiuseppe CasalicchioMarvin N. WrightBernd BischlPublished in: CoRR (2021)
Keyphrases
- data sets
- input data
- prior knowledge
- image data
- data sources
- data analysis
- redundant data
- database
- data processing
- synthetic data
- process model
- machine learning
- data collection
- feature importance
- training data
- missing data
- test set
- semi supervised learning
- k nearest neighbor
- information retrieval
- learning algorithm
- similarity measure
- data streams
- data structure