Letter Level Learning for Language Independent Diacritics Restoration

Description:

This paper discusses letter level learning for language independent diacritics restoration.

Creator(s):
Creation Date: September 2002
Partner(s):
UNT College of Engineering
Collection(s):
UNT Scholarly Works
Usage:
Total Uses: 41
Past 30 days: 4
Yesterday: 0
Creator (Author):
Mihalcea, Rada, 1974-

University of North Texas

Creator (Author):
Nastase, Vivi

University of Ottawa

Date(s):
  • Creation: September 2002
Description:

This paper discusses letter level learning for language independent diacritics restoration.

Degree:
Note:

Abstract: This paper presents a method for diacritics restoration based on learning mechanisms that act at letter level. The method requires no additional tagging tools or resources other than raw text, which makes it independent of the language, and particularly appealing for languages for which there are few resources available. The algorithm was evaluated on four different languages, namely Czech, Hungarian, Polish, and Romanian, and an average accuracy of over 98% was observed.

Physical Description:

7 p.

Language(s):
Subject(s):
Keyword(s): diacritics restorations | languages
Source: Sixth Conference on Natural Language Learning (CoNLL), 2002, Taipei, Taiwan
Partner:
UNT College of Engineering
Collection:
UNT Scholarly Works
Identifier:
  • ARK: ark:/67531/metadc30944
Resource Type: Paper
Format: Text
Rights:
Access: Public