A Multilingual Information Extraction Pipeline for Investigative Journalism.
Gregor WiedemannSeid Muhie YimamChris BiemannPublished in: EMNLP (Demonstration) (2018)
Keyphrases
- information extraction
- natural language processing
- precision and recall
- text mining
- free text
- web documents
- question answering
- digital libraries
- structured data
- named entities
- cross language
- web mining
- relation extraction
- multi lingual
- textual data
- open domain
- information retrieval
- machine learning
- language independent
- semantic tagging
- text documents
- named entity recognition
- semi structured
- machine translation
- relational learning
- conditional random fields
- natural language
- information extraction systems
- cross language information retrieval
- text processing
- cross lingual
- data extraction
- pipeline architecture
- extracting meaningful
- database
- multilingual information retrieval
- processing pipeline
- language specific
- unstructured text
- natural language text