Research output: Contribution to journal › Conference article › peer-review
Tagging Russian texts of the XIXth century has been evaluated. The causes have been determined why some words turned out to be unknown to the tagger, i.e. remained without lemmas and grammatical features. The investigation showed that the main reasons of the existence of the unknown words were as follows: 1) incompleteness of the tagger dictionary, particularly in the XIXth century lexical stock; 2) failure to tag the word-formative derivates; 3) problems with some inflexion models of Old Russian; 4) insufficiency of graphemic analysis; 5) inability of taggers to process multiwords. The results obtained provide a baseline to improve premorphological processing of Russian texts and to work out the more sophisticated approaches to morphological analysis.
Original language | English |
---|---|
Pages (from-to) | 235-242 |
Number of pages | 8 |
Journal | Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) |
Volume | 3206 |
State | Published - 1 Dec 2004 |
Event | 7th International Conference TSD 2004: Text, Speech and Dialogue - Brno, Czech Republic Duration: 8 Sep 2004 → 11 Sep 2004 |
ID: 30268644