Unsupervised deep embedding for novel class detection over data stream.
Ahmad M. MustafaGbadebo AyoadeKhaled Al-NaamiLatifur KhanKevin W. HamlenBhavani M. ThuraisinghamFrederico AraujoPublished in: IEEE BigData (2017)
Keyphrases
- data streams
- concept drift
- data stream classification
- sliding window
- deep architectures
- streaming data
- semi supervised
- change detection
- data stream mining
- transactional data
- unsupervised learning
- space efficient
- distributed data streams
- deep learning
- data sets
- stream data
- evolving data streams
- data distribution
- supervised learning
- stream mining
- vector space
- graph embedding
- high level declarative
- stream processing
- machine learning
- unsupervised manner
- weakly supervised
- incoming data
- data hiding
- continuous queries
- itemsets
- continuous stream
- heavy hitters
- continuous data streams
- digital images