Results 1 -
5 of
5
Improvements in analogical learning: application to translating multi-terms of the medical domain
- In Proceedings of the 12th Conference of the European Chapter of the ACL (EACL, 2009
"... Handling terminology is an important matter in a translation workflow. However, current Machine Translation (MT) systems do not yet propose anything proactive upon tools which assist in managing terminological databases. In this work, we investigate several enhancements to analogical learning and te ..."
Abstract
-
Cited by 8 (0 self)
- Add to MetaCart
(Show Context)
Handling terminology is an important matter in a translation workflow. However, current Machine Translation (MT) systems do not yet propose anything proactive upon tools which assist in managing terminological databases. In this work, we investigate several enhancements to analogical learning and test our implementation on translating medical terms. We show that the analogical engine works equally well when translating from and into a morphologically rich language, or when dealing with language pairs written in different scripts. Combining it with a phrasebased statistical engine leads to significant improvements. 1
Mitigating Problems in Analogy-based EBMT with SMT and vice versa: a Case Study with Named Entity Transliteration∗
"... Abstract. Five years ago, a number of papers reported an experimental implementation of an Example Based Machine Translation (EBMT) system using proportional analogy. This approach, a type of analogical learning, was attractive because of its simplicity; and the paper reported considerable success w ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
Abstract. Five years ago, a number of papers reported an experimental implementation of an Example Based Machine Translation (EBMT) system using proportional analogy. This approach, a type of analogical learning, was attractive because of its simplicity; and the paper reported considerable success with the method using various language pairs. In this paper, we describe our attempt to use this approach for tackling English–Hindi Named Entity (NE) Transliteration. We have implemented our own EBMT system using proportional analogy and have found that the analogy-based system on its own has low precision but a high recall due to the fact that a large number of names are untransliterated with the approach. However, mitigating problems in analogy-based EBMT with SMT and vice-versa have shown considerable improvement over the individual approach.
PACLIC 24 Proceedings 365 Mitigating Problems in Analogy-based EBMT with SMT and vice versa: a Case Study with Named Entity Transliteration∗
"... Abstract. Five years ago, a number of papers reported an experimental implementation of an Example Based Machine Translation (EBMT) system using proportional analogy. This approach, a type of analogical learning, was attractive because of its simplicity; and the paper reported considerable success w ..."
Abstract
- Add to MetaCart
Abstract. Five years ago, a number of papers reported an experimental implementation of an Example Based Machine Translation (EBMT) system using proportional analogy. This approach, a type of analogical learning, was attractive because of its simplicity; and the paper reported considerable success with the method using various language pairs. In this paper, we describe our attempt to use this approach for tackling English–Hindi Named Entity (NE) Transliteration. We have implemented our own EBMT system using proportional analogy and have found that the analogy-based system on its own has low precision but a high recall due to the fact that a large number of names are untransliterated with the approach. However, mitigating problems in analogy-based EBMT with SMT and vice-versa have shown considerable improvement over the individual approach.
Mapping Source to Target Strings without Alignment by Analogical Learning: A Case Study with Transliteration
"... Analogical learning over strings is a holistic model that has been investigated by a few authors as a means to map forms of a source language to forms of a target language. In this study, we revisit this learning paradigm and apply it to the transliteration task. We show that alone, it performs wors ..."
Abstract
- Add to MetaCart
(Show Context)
Analogical learning over strings is a holistic model that has been investigated by a few authors as a means to map forms of a source language to forms of a target language. In this study, we revisit this learning paradigm and apply it to the transliteration task. We show that alone, it performs worse than a statistical phrase-based machine translation engine, but the combination of both approaches outperforms each one taken separately, demonstrating the usefulness of the information captured by a so-called formal analogy. 1
Towards automatic acquisition of linguistic features
"... This paper proposes a method to acquire linguistic features from a corpus of short sentences by extracting analogous sen-tences like what ’s the next station?: where ’s the bus station?:: what is the next stop? : where is the bus stop? The procedures used to construct clusters of analogous sentences ..."
Abstract
- Add to MetaCart
(Show Context)
This paper proposes a method to acquire linguistic features from a corpus of short sentences by extracting analogous sen-tences like what ’s the next station?: where ’s the bus station?:: what is the next stop? : where is the bus stop? The procedures used to construct clusters of analogous sentences are presented. Exper-iments performed on roughly 40,000 short sentences from the tourism domain in En-glish and Japanese are reported, and the clusters produced are analyzed and inter-preted in terms of linguistic features. 1
