Deconstruct to Reconstruct a Configurable Evaluation Metric for Open-Domain Dialogue Systems.
Vitou PhyYang ZhaoAkiko AizawaPublished in: CoRR (2020)
Keyphrases
- open domain
- evaluation metrics
- dialogue system
- information extraction
- precision and recall
- natural language
- question answering
- average precision
- tutorial dialogue
- spoken dialogue systems
- evaluation measures
- question answering systems
- natural language processing
- learning to rank
- human users
- user model
- machine learning
- data mining
- loss function
- reinforcement learning
- information retrieval