Iterative Refinement and Quality Checking of Annotation Guidelines - How to Deal Effectively with Semantically Sloppy Named Entity Types, such as Pathological Phenomena.
Udo HahnElena BeisswangerEkaterina BuykoErik FaesslerJenny TraumüllerSusann SchröderKerstin HornbostelPublished in: LREC (2012)
Keyphrases
- named entities
- iterative refinement
- annotated corpus
- named entity recognition
- information extraction
- named entity extraction
- co occurrence
- natural language processing
- relation extraction
- question answering
- person names
- automatic annotation
- weakly supervised
- text mining
- noun phrases
- active learning
- image retrieval
- natural language
- maximum entropy model
- data sets
- proper names
- unsupervised learning
- image annotation
- semantic annotation
- dimensionality reduction
- semi supervised
- pattern recognition
- image segmentation
- decision trees
- metadata
- databases