Can Word Probabilities from LDA be Simply Added up to Represent Documents?
Zhiqiang CaiHaiying LiXiangen HuArt GraesserPublished in: EDM (2016)
Keyphrases
- latent topics
- latent dirichlet allocation
- topic models
- topic modeling
- word spotting
- latent semantic analysis
- topic discovery
- text documents
- statistical topic models
- keywords
- information retrieval systems
- information retrieval
- face recognition
- document collections
- co occurrence
- latent dirichlet
- word frequencies
- linear discriminant analysis
- word pairs
- natural language text
- probabilistic topic models
- printed documents
- linguistic information
- text corpus
- lda model
- sentence level
- term frequency
- page layout
- text analysis
- text mining
- document representation
- principal component analysis
- generative model
- concept space
- relevant documents
- related words
- related documents
- multiword
- vector space model
- document retrieval
- bag of words
- word frequency
- web documents
- dimensionality reduction
- language model
- document space
- feature extraction
- n gram