Identifying predictive hubs to condense the training set of (k)-nearest neighbour classifiers.
Ludwig LausserChristoph MüsselAlexander MelkozerovHans A. KestlerPublished in: Comput. Stat. (2014)
Keyphrases
- k nearest neighbour
- training set
- nearest neighbour
- knn
- support vector machine
- nearest neighbor
- k nearest neighbor
- classification algorithm
- information gain
- training data
- test set
- classification accuracy
- text categorisation
- decision trees
- supervised learning
- training samples
- svm classifier
- active learning
- multi layer perceptron
- data sets
- test data
- machine learning
- naive bayes
- decision boundary
- neural network
- distance function
- similarity search
- feature selection
- base classifiers
- data mining
- pattern recognition