Lasige-BioTM at ProfNER: BiLSTM-CRF and contextual Spanish embeddings for Named Entity Recognition and Tweet Binary Classification.
Pedro RuasVitor D. T. AndradeFrancisco M. CoutoPublished in: SMM4H@NAACL-HLT (2021)
Keyphrases
- named entity recognition
- binary classification
- conditional random fields
- named entities
- question answering
- information extraction
- multi class
- support vector
- cost sensitive
- natural language processing
- learning problems
- generalization error
- multi label
- crf model
- prediction accuracy
- text summarization
- support vector machine
- maximum entropy
- graphical models
- hidden markov models
- pairwise
- semi supervised
- ensemble methods
- class imbalance
- co occurrence
- higher order
- classifier ensemble
- markov random field
- probabilistic model
- dimensionality reduction
- text mining
- active learning
- generative model
- machine learning
- training data
- computer vision
- learning tasks
- model selection
- learning environment
- image processing
- information retrieval