Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction.
Chen LinTimothy A. MillerDmitriy DligachHadi AmiriSteven BethardGuergana SavovaPublished in: Louhi@EMNLP (2018)
Keyphrases
- recurrent neural networks
- relation extraction
- automatic extraction
- information extraction
- named entities
- domain specific
- semantic relations
- semantic features
- neural network
- semantic role labeling
- broad coverage
- question answering
- named entity recognition
- recurrent networks
- feed forward
- echo state networks
- dependency trees
- artificial neural networks
- semi supervised
- training set
- parse tree
- tree kernels
- artificial intelligence
- domain knowledge
- information retrieval
- structured data
- co occurrence
- general purpose
- knowledge representation
- genetic algorithm