Quootstrap: Scalable Unsupervised Extraction of Quotation-Speaker Pairs from Large News Corpora via Bootstrapping.
Dario PavlloTiziano PiccardiRobert WestPublished in: ICWSM (2018)
Keyphrases
- information extraction
- completely unsupervised
- natural language processing
- weakly supervised
- unsupervised learning
- pairwise
- semi supervised
- automatic extraction
- social media
- news corpus
- unsupervised methods
- named entity recognition
- data driven
- speech recognition
- supervised learning
- linguistic patterns
- relation extraction
- machine learning
- unsupervised manner
- broadcast news
- online news
- speaker recognition
- story segmentation
- speaker diarization
- bilingual lexicon
- speaker verification
- automatic speech recognition
- news stories
- audio visual
- news articles
- hidden markov models