Protein named entity classification with probabilistic features derived from GENIA corpus and MEDLINE.
Sagara SumathipalaKoichi YamadaMuneyuki UneharaPublished in: SCIS&ISIS (2014)
Keyphrases
- named entities
- genia corpus
- text mining
- maximum entropy model
- classification accuracy
- feature extraction
- feature set
- co occurrence
- feature vectors
- feature space
- information extraction
- unsupervised learning
- text classification
- global context
- contextual features
- question answering
- relation extraction
- pattern recognition
- named entity recognition
- named entity extraction
- class labels
- natural language processing
- decision trees
- artificial intelligence
- annotated corpus
- image classification
- proper names
- linguistic features
- noun phrases
- machine learning
- supervised learning
- training data
- weakly supervised
- training set
- bayesian networks
- information retrieval