A benchmark of classifiers on feature drifting data streams.
Jean Paul BarddalHeitor Murilo GomesAlceu de Souza Britto Jr.Fabrício EnembreckPublished in: ICPR (2016)
Keyphrases
- data streams
- concept drift
- data stream classification
- feature set
- anytime classification
- feature weights
- classification algorithm
- streaming data
- sliding window
- real world
- ensemble classifier
- mining concept drifting data streams
- classification models
- sensor data
- feature selection
- decision trees
- image features
- naive bayes
- multiple classifiers
- stream data
- feature values
- support vector
- data sets
- multiple features
- test set
- input features
- feature ranking
- data stream mining
- feature vectors
- classifier combination
- machine learning
- adaboost classifier
- image classification
- single feature
- sensor networks
- machine learning methods
- change detection
- training data
- supervised classification
- feature subset