We present a novel method for discovering parallel sentences in comparable, non-parallel corpora. We train a maximum entropy classifier that, given a pair of sentences, can reliably determine whether or not they are translations of each other. Using this approach, we extract parallel data from large Chinese, Arabic, and English non-parallel newspaper corpora. We evaluate the quality of the extracted data by showing that it improves the performance of a state-of-the-art statistical machine translation system. We also show that a good-quality MT system can be built from scratch by starting with a very small parallel corpus (100,000 words) and exploiting a large non-parallel corpus. Thus, our method can be applied with great benefit to language pairs for which only scarce resources are available. 1.

We investigate a number of simple methods for improving the word-alignment accuracy of IBM Model 1. We demonstrate reduction in alignment error rate of approximately 30 % resulting from (1) giving extra weight to the probability of alignment to the null word, (2) smoothing probability estimates for rare words, and (3) using a simple heuristic estimation method to initialize, or replace, EM training of model parameters.

Increased availability of parallel data and recent progress in modeling, decoding, and evaluation have recently had a major impact on machine translation (MT) accuracy. This paper covers the basic elements of state-of-the-art, statistical MT. 1.

Semantic parsing is the construction of a complete, formal, symbolic meaning representation of a sentence. While it is crucial to natural language understanding, the problem of semantic parsing has received relatively little attention from the machine learning community. Recent work on natural language understanding has mainly focused on shallow semantic analysis, such as word-sense disambiguation and semantic role labeling. Semantic parsing, on the other hand, involves deep semantic analysis in which word senses, semantic roles and other components are combined to produce useful meaning representations for a particular application domain (e.g. database query). Prior research in machine learning for semantic parsing is mainly based on inductive logic programming or deterministic parsing, which lack some of the robustness that characterizes statistical learning. Existing statistical approaches to semantic parsing, however, are mostly concerned with relatively simple application domains in which a meaning representation is no more than a single semantic frame. In this proposal, we present a novel statistical approach to semantic parsing, WASP, which can handle meaning representations with a nested structure. The WASP algorithm learns a semantic parser given a set of sentences annotated with their correct meaning representations. The parsing model is based on the

Abstract. For many applications such as machine translation and bilingual information retrieval, the bilingual corpora play an important role in training the system. Because they are obtained through automatic or semi automatic methods, they usually include noise, sentence pairs which are worthless or even harmful for training the system. We study the effect of different levels of corpus noise on an end-to-end statistical machine translation system. We also propose an efficient method for corpus filtering. This method filters out the noisy part of a corpus based on the state-of-the-art word alignment models. We show the efficiency of this method on the basis of the sentence misalignment rate of the filtered corpus and its positive effect on the translation quality. 1

Abstract. Parallel corpora are playing a crucial role in multilingual natural language processing. Unfortunately, the availability of such a resource is the bottleneck in most applications of interest. Mining the web for parallel corpora is a viable solution that comes at a price: it is not always easy to identify parallel documents among the crawled material. In this study we address the problem of automatically identifying the pairs of texts that are translation of each other in a set of documents. We show that it is possible to automatically build particularly efficient content-based methods that make use of very little lexical knowledge. We also evaluate our approach toward a front-end translation task and demonstrate that our parallel text classifier yields better performances than another approach based on a rich lexicon. 1

Abstract not found

Lack of sufficient linguistic resources and parallel corpora for many languages and domains currently is one of the major obstacles to further advancement of automated translation. The solution proposed in this paper is to exploit the fact that non-parallel bi- or multilingual text resources are much more widely available than parallel translation data. This position paper presents previous research in this field and research plans of the ACCURAT project. Its goal is to find, analyze and evaluate novel methods that exploit comparable corpora in order to compensate for the shortage of linguistic resources, and ultimately to significantly improve MT quality for under-resourced languages and narrow domains. 1.

While several recent works on dealing with large bilingual collections of texts, e.g. (Smith et al., 2010), seek for extracting parallel sentences from comparable corpora, we present PARADOCS, a system designed to recognize pairs of parallel documents in a (large) bilingual collection of texts. We show that this system outperforms a fair baseline (Enright and Kondrak, 2007) in a number of controlled tasks. We applied it on the French-English cross-language linked article pairs of Wikipedia in order see whether parallel articles in this resource are available, and if our system is able to locate them. According to some manual evaluation we conducted, a fourth of the article pairs in Wikipedia are indeed in translation relation, and PARADOCS identifies parallel or noisy parallel article pairs with a precision of 80%. 1

Abstract not found