Abstract
I propose a text normalization model based on learning edit operations from labeled data while incorporating features induced from unlabeled text and from dictionaries. These features enable effective learning with little supervision, as demonstrated on an English tweet normalization dataset.
Original language | English |
---|---|
Publication status | Published - 2014 |
Event | ATILA 2014 - Ghent, Belgium Duration: 20 Nov 2014 → 21 Nov 2014 |
Conference
Conference | ATILA 2014 |
---|---|
Country/Territory | Belgium |
City | Ghent |
Period | 20/11/14 → 21/11/14 |