Header

UZH-Logo

Maintenance Infos

Enhancing a Rule-Based MT System with Cross-Lingual WSD


Rudnick, Alex; Rios, Annette; Gasser, Michael (2014). Enhancing a Rule-Based MT System with Cross-Lingual WSD. In: SaLTMiL Workshop on free/open-source language resources for the machine translation of less-resourced languages (LREC'14), Reykjavik, Iceland, 22 May 2014 - 22 May 2014, 31-36.

Abstract

Lexical ambiguity is a significant problem facing rule-based machine translation systems, as many words have several possible translations in a given target language, each of which can be considered a sense of the word from the source language. The difficulty of resolving these ambiguities is mitigated for statistical machine translation systems for language pairs with large bilingual corpora, as large n-gram language models and phrase tables containing common multi-word expressions can encourage coherent word choices.
For most language pairs these resources are not available, so a primarily rule-based approach becomes attractive. In cases where some training data is available, though, we can investigate hybrid RBMT and machine learning approaches, leveraging small and potentially growing bilingual corpora. In this paper we describe the integration of statistical cross-lingual word-sense disambiguation software with SQUOIA, an existing rule-based MT system for the Spanish-Quechua language pair, and show how it allows us to learn from the available bitext to make better lexical choices, with very few code changes to the base system. We also describe Chipa, the new open source CL-WSD software used for these experiments.

Abstract

Lexical ambiguity is a significant problem facing rule-based machine translation systems, as many words have several possible translations in a given target language, each of which can be considered a sense of the word from the source language. The difficulty of resolving these ambiguities is mitigated for statistical machine translation systems for language pairs with large bilingual corpora, as large n-gram language models and phrase tables containing common multi-word expressions can encourage coherent word choices.
For most language pairs these resources are not available, so a primarily rule-based approach becomes attractive. In cases where some training data is available, though, we can investigate hybrid RBMT and machine learning approaches, leveraging small and potentially growing bilingual corpora. In this paper we describe the integration of statistical cross-lingual word-sense disambiguation software with SQUOIA, an existing rule-based MT system for the Spanish-Quechua language pair, and show how it allows us to learn from the available bitext to make better lexical choices, with very few code changes to the base system. We also describe Chipa, the new open source CL-WSD software used for these experiments.

Statistics

Downloads

106 downloads since deposited on 29 Jul 2014
49 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Paper), not refereed, original work
Communities & Collections:06 Faculty of Arts > Institute of Computational Linguistics
Dewey Decimal Classification:000 Computer science, knowledge & systems
410 Linguistics
Uncontrolled Keywords:under-resourced languages, hybrid machine translation, word-sense disambiguation
Language:English
Event End Date:22 May 2014
Deposited On:29 Jul 2014 10:44
Last Modified:05 Apr 2016 17:59
Publisher:SALTMIL
Official URL:http://ixa2.si.ehu.es/saltmil/
Related URLs:http://siuc01.si.ehu.es/~jipsagak/SALTMIL/LREC_2014_Workshop_Proceedings_Saltmil.pdf

Download

Preview Icon on Download
Preview
Filetype: PDF
Size: 150kB