Papers by Siim Orasmaa
Named Entity Recognition in Estonian 19th Century Parish Court Records (2022.lrec-1)
Copied to clipboard
| Challenge: | a corpus of 19th century Parish Court records annotated for named entities (NE) in Estonian is a valuable resource for historians, linguists and the public at large. |
| Approach: | They propose to annotate a corpus of Estonian Parish Court records annotated for named entities (NE) and report on named entity recognition experiments using this corpus. |
| Outcome: | The proposed model achieves microaverage F1 score of 93.6, comparable to state-of-the-art NER performance on the contemporary Estonian. |
EstNLTK 1.6: Remastered Estonian NLP Pipeline (2020.lrec-1)
Copied to clipboard
| Challenge: | EstNLTK Python library contains a unified programming interface for natural language processing in Estonian. |
| Approach: | They propose to provide a unified programming interface for natural language processing in Estonian while preserving the intuitive interface for novices. |
| Outcome: | The new interface greatly simplifies processing of large text collections. |