An approach to dealing with missing values in heterogeneous data using k-nearest neighbors.
Davi E. N. FrossardIgor O. NunesRenato A. KrohlingPublished in: CoRR (2016)
Keyphrases
- missing values
- heterogeneous data
- k nearest neighbor
- high dimensional data
- nearest neighbor
- knn
- complex data
- missing data
- data integration
- data management
- incomplete data
- data sources
- metadata
- data imputation
- support vector machine
- databases
- neural network
- distance function
- k nearest neighbour
- high dimensional
- dimensionality reduction
- information sources
- imputation methods
- feature selection
- text classification
- classification algorithm
- training set
- similarity search
- data model
- business intelligence
- feature extraction
- web data
- learning algorithm
- test instances
- data points