From Tokenization to Self-Supervision: Building a High-Performance Information Extraction System for Chemical Reactions in Patents.
Jingqi WangYuankai RenZhi ZhangHua XuYaoyun ZhangPublished in: Frontiers Res. Metrics Anal. (2021)
Keyphrases
- information extraction
- chemical reactions
- named entities
- information retrieval
- chemical reaction
- natural language processing
- text mining
- prior art
- extracting meaningful
- relation extraction
- free text
- active learning
- natural language
- open domain
- textual data
- relational learning
- united states
- high reliability
- web documents
- machine learning
- biomedical text
- query expansion
- scientific literature
- named entity recognition
- structured data
- precision and recall
- semi structured
- web mining
- machine translation