This paper presents a method for exploiting document-level similarity between the documents in the training corpus for a corpus-driven (statistical or example-based) machine translation system and the input documents it must translate. The method is simple to implement, efficient (increases the translation time of an example-based system by only a few percent), and robust (still works even when the actual document boundaries in the input text are not known). Experiments on French-English and Arabic-English showed relative gains over the same system without using document-level similarity of up to 7.4% and 5.4%, respectively, on the BLEU metric.

Example-Based Machine Translation (EBMT) systems have typically operated on individual sentences without taking into account prior context. By adding a simple reweighting of retrieved fragments of training examples on the basis of whether the previous translation retrieved any fragments from examples within a small window of the current instance, translation performance is improved. A further improvement is seen by performing a similar reweighting when another fragment of the current input sentence was retrieved from the same training example. Together, a simple, straightforward implementation of these two factors results in an improvement on the order of 1.0–1.6% in the BLEU metric across multiple data sets in multiple languages.

The Burrows-Wheeler Transform (BWT) was originally developed for data compression, but can also be applied to indexing text. In this paper, an adaptation of the BWT to word-based indexing of the training corpus for an example-based machine translation (EBMT) system is presented. The adapted BWT embeds the necessary information to retrieve matched training instances without requiring any additional space and can be instantiated in a compressed form which reduces disk space and memory requirements by about 40% while still remaining searchable without decompression. Both the speed advantage from O(log N) lookups compared to the O(N) lookups in the inverted-file index which had previously been used and the structure of the index itself act as enablers for additional capabilities and run-time speed. Because the BWT groups all instances of any n-gram together, it can be used to quickly enumerate the most-frequent n-grams, for which translations can be precomputed and stored, resulting in an order-of-magnitude speedup at run time.

Many corpus-based Machine Translation (MT) systems generate a number of partial translations which are then pieced together rather than immediately producing one overall translation. While this makes them more robust to ill-formed input, they are subject to disfluencies at phrasal translation boundaries even for well-formed input. We address this “boundary friction” problem by introducing a method that exploits overlapping phrasal translations and the increased confidence in translation accuracy they imply. We specify an efficient algorithm for producing translations using overlap. Finally, our empirical analysis indicates that this approach produces higher quality translations than the standard method of combining non-overlapping fragments generated by our Example-Based MT (EBMT) system in a peak-to-peak comparison.

Machine Translation of minority languages presents unique challenges, including the paucity of bilingual training data and the unavailability of linguistically-trained speakers. This paper focuses on a machine learning approach to transfer-based MT, where data in the form of translations and lexical alignments are elicited from bilingual speakers, and a seeded version-space learning algorithm formulates and refines transfer rules. A rule-generalization lattice is defined based on LFG-style f-structures, permitting generalization operators in the search for the most general rules consistent with the elicited data. The paper presents these methods and illustrates examples.

Pre-processing of bilingual corpora plays an important role in Example-Based Machine Translation (EBMT) and Statistical-Based Machine Translation (SBMT). For our Mandarin-English EBMT system, pre-processing includes segmentation for Mandarin, bracketing for English and building a statistical dictionary from the corpora. We used the Mandarin segmenter from the Linguistic Data Consortium (LDC). It uses dynamic programming with a frequency dictionary to segment the text. Although the frequency dictionary is large, it does not completely cover the corpora. In this paper, we describe the work we have done to improve the segmentation for Mandarin and the bracketing process for English to increase the length of English phrases. A statistical dictionary is built from the aligned bilingual corpus. It is used as feedback to segmentation and bracketing to re-segment / re-bracket the corpus. The process iterates several times to achieve better results. The final results of the corpus pre-processing are a segmented/bracketed aligned bilingual corpus and a statistical dictionary. We achieved positive results by increasing the average length of Chinese terms about 60% and 10% for English. The statistical dictionary gained about a 30% increase in coverage.

NICE is a machine translation project for low-density languages. We are building a tool that will elicit a controlled corpus from a bilingual speaker who is not an expert in linguistics. The corpus is intended to cover major typological phenomena, as it is designed to work for any language. Using implicational universals, we strive to minimize the number of sentences that each informant has to translate. From the elicited sentences, we learn transfer rules with a version space algorithm. Our vision for MT in the future is one in which systems can be quickly trained for new languages by native speakers, so that speakers of minor languages can participate in education, health care, government, and internet without having to give up their languages.