Automatic Evaluation and Uniform Filter Cascades for Inducing N-Best Translation Lexicons

Automatic Evaluation and Uniform Filter Cascades for Inducing N-Best Translation Lexicons I. Dan Melamed Department of Computer and Information Science University of Pennsylvania Philadelphia, PA, 19104 U.S.A. melamed~unagi, cis. upenn, edu Abstract This paper shows how to induce an N-best translation lexicon from a bilingual text corpus using statistical properties of the corpus together with four external knowledge sources. The knowledge sources are cast as filters, so that any subset of them can be cascaded in a uniform framework. A new objective evaluation measure is used to compare the quality of lexicons induced with different filter cascades. The best filter cascades improve lexicon quality by up to 137% over the plain vanilla statistical method, and approach human performance. Drastically reducing the size of the training corpus has a much smaller impact on lexicon quality when these knowledge sources are used. This makes it practical to train on small hand-built corpora for language pairs where large bilingual corpora are unavailable. Moreover, three of the four filters prove useful even when used with large training corpora. 1 INTRODUCTION A machine translation system must be able to choose among possible translations based on context. To do this, it usuMly relies on a translation lexicon that contains a number of possible translations for each word. N-best translation lexicons contain up to N candidate translations for each word, or- dered from most probable to least probable, sometimes specifying a priori probabilities or likelihood scores. Existing automatic methods for constructing N-best translation lexicons rely on the availability of large training corpora of parallel texts in the source and target languages. For some methods, the corpora must also be aligned by sentence [Bro93, Ga191a]. Unfortunately, such training corpora are available for only a handful of language pairs, and the cost to create enough training data manually for new language pairs is very high. This paper presents 1. a new automatic evaluation method for N-best translation lexicons, 2. a filter-based approach for enhancing statistical translation models with non-statistical sources of information, 3. four sources of information that can drastically reduce the necessary amount of training material. 184 I The evaluation method uses a simple objective criterion rather than relying on subjective human judges. It allows many experiments to be run without concern about the cost, availability and reliability of human evaluators. The filter-based approach is designed to identify likely (source word, target word) 1 pairs, using a statistical decision procedure. Candidate word pairs are drawn from a corpus of aligned sentences: (S, T) is a candidate iff T appears in the translation of a sentence containing S. In the simplest case, the decision procedure considers M1 candidates for inclusion in the lexicon; but the new framework allows a cascade of non-statistical filters to remove inappropriate pairs fl'om consideration. Each filter is based on a particular knowledge source, and can be placed into the cascade independently of the others. The knowledge sources investigated here are: • part of speech information, • machine-readable bilingual dictionaries (MRBDs), • cognate heuristics, and …

Content truncated to 500 words