Improving NCD accuracy by combining document segmentation and document distortion.
Ana GranadosRafael MartínezDavid CamachoFrancisco de Borja RodríguezPublished in: Knowl. Inf. Syst. (2014)
Keyphrases
- information retrieval systems
- web documents
- document classification
- document collections
- information retrieval
- document clustering
- document images
- multiscale
- fully automatic
- page segmentation
- tf idf
- text documents
- segmentation algorithm
- high accuracy
- document content
- segmentation method
- prediction accuracy
- relevant documents
- level set
- region growing
- computational complexity
- video sequences
- document representation
- segmentation accuracy
- document analysis
- decision trees