Identifying Document Metadata Based on Multilayer Clustering.
Joris D'hondtDennis VandevennePaul-Armand VerhaegenJoris VertommenDirk CattrysseJoost R. DuflouPublished in: DET (2009)
Keyphrases
- metadata
- document clustering
- digital documents
- clustering algorithm
- digital libraries
- multimedia documents
- document collections
- tolerance rough set
- clustering method
- text clustering
- cluster membership
- data clustering
- document images
- k means
- topic discovery
- cluster analysis
- database
- semantic information
- hierarchical clustering
- spectral clustering
- self organizing maps
- unsupervised learning
- dublin core
- keywords
- information retrieval
- web documents
- document retrieval
- information resources
- document classification
- retrieval systems
- document analysis
- cosine similarity
- learning objects
- co occurrence
- web objects
- document clusters
- text mining
- information extraction
- similarity measure