Mining Drifting Data Streams on a Budget: Combining Active Learning with Self-Labeling.
Lukasz KoryckiBartosz KrawczykPublished in: CoRR (2021)
Keyphrases
- active learning
- data streams
- concept drift
- stream mining
- continuous data streams
- data stream mining
- multiple data streams
- itemsets
- mining data streams
- frequent itemsets
- sliding window
- streaming data
- closed frequent itemsets
- learning algorithm
- active learning strategies
- limited memory
- learning strategies
- stream data
- labeled data
- random sampling
- machine learning
- sequential patterns
- sample selection
- training set
- mining algorithm
- data sets
- experimental design
- labeling effort
- sensor data
- semi supervised learning
- high speed data streams
- relevance feedback
- change detection
- cost sensitive
- learning process
- unlabeled data
- semi supervised
- data mining techniques
- selective sampling
- data mining
- concept drifting
- frequent closed itemsets
- pool based active learning
- web mining
- pattern mining
- data distribution
- supervised learning
- knowledge discovery
- data structure
- stream processing
- outlier detection
- frequent patterns
- training examples
- text mining
- sensor networks
- noisy data streams