Tags, Titles or Q&As?: Choosing Content Descriptors for Visual Recommender Systems.
Belgin MutluEduardo E. VeasChristoph TrattnerPublished in: HT (2017)
Keyphrases
- user profiling
- recommender systems
- social tags
- user interests
- user profiles
- visual similarity
- metadata
- collaborative filtering
- user generated
- web images
- personalized recommendation
- textual content
- visual information
- visual features
- information overload
- music recommendation
- user model
- visual appearance
- part of speech
- keywords
- semantic content
- visual concepts
- visual descriptors
- web resources
- semantic tags
- category recognition
- active user
- user contributed
- concept detectors
- multimedia
- user generated content
- user preferences
- textual descriptions
- video segments
- image descriptors
- user modeling
- low level
- image retrieval
- social tagging
- visual representations
- cold start
- web content
- web documents