MMDF-LDA: An improved Multi-Modal Latent Dirichlet Allocation model for social image annotation.
Liu ZhengCaiming ZhangCaixian ChenPublished in: Expert Syst. Appl. (2018)
Keyphrases
- multi modal
- latent dirichlet allocation
- image annotation
- lda model
- topic models
- probabilistic latent semantic analysis
- topic modeling
- hierarchical bayesian model
- probabilistic model
- variational inference
- gibbs sampling
- high level
- semantic concepts
- image understanding
- latent topics
- information retrieval
- generative model
- image content
- automatic image annotation
- text classification
- feature set
- text mining
- natural language processing
- prior knowledge
- image retrieval
- latent topic models
- variational bayesian inference