Naively collecting translations by crowdsourcing the task to non-professional translators yields disfluent, low-quality results if no quality control is exercised. We demonstrate a variety of mechanisms that increase the translation quality to near professional levels. Specifically, we solicit redundant translations and edits to them, and automatically select the best output among them. We propose a set of features that model both the translations and the translators, such as country of residence, LM perplexity of the translation, edit rate from the other translations, and (optionally) calibration against professional translators. Using these features to score the collected translations, we are able to discriminate between acceptable and unacceptable translations. We recreate the NIST 2009 Urdu-to-English evaluation set with Mechanical Turk, and quantitatively show that our models are able to select translations within the range of quality that we expect from professional translators. The total cost is more than an order of magnitude lower than professional translation. 1

We estimate the parameters of a phrasebased statistical machine translation system from monolingual corpora instead of a bilingual parallel corpus. We extend existing research on bilingual lexicon induction to estimate both lexical and phrasal translation probabilities for MT-scale phrasetables. We propose a novel algorithm to estimate reordering probabilities from monolingual data. We report translation results for an end-to-end translation system using these monolingual features alone. Our method only requires monolingual corpora in source and target languages, a small bilingual dictionary, and a small bitext for tuning feature weights. In this paper, we examine an idealization where a phrase-table is given. We examine the degradation in translation performance when bilingually estimated translation probabilities are removed and show that 80%+ of the loss can be recovered with monolingually estimated features alone. We further show that our monolingual features add 1.5 BLEU points when combined with standard bilingually estimated phrase table features. 1

Parallel text acquisition from the Web is an attractive way for augmenting statistical models (e.g., machine translation, crosslingual document retrieval) with domain representative data. The basis for obtaining such data is a collection of pairs of bilingual Web sites or pages. In this work, we propose a crawling strategy that locates bilingual Web sites by constraining the visitation policy of the crawler to the graph neighborhood of bilingual sites on the Web. Subsequently, we use a novel recursive mining technique that recursively extracts text and links from the collection of bilingual Web sites obtained from the crawling. Our method does not suffer from the computationally prohibitive combinatorial matching typically used in previous work that uses document retrieval techniques to match a collection of bilingual webpages. We demonstrate the efficacy of our approach in the context of machine translation in the tourism and hospitality domain. The parallel text obtained using our novel crawling strategy results in a relative improvement of 21 % in BLEU score (English-to-Spanish) over an out-of-domain seed translation model trained on the European parliamentary proceedings. Index Terms: Web crawling, parallel text, machine translation 1.

This paper presents a novel crawling strategy to locate bilingual sites. It does so by focusing on the Web graph neighborhood of these sites and exploring the patterns of the links in this region to guide its visitation policy. A sub-task in the problem of bilingual site discovery is the job of detecting bilingual sites, i.e., given a Web site, verify whether it is bilingual or not. We perform this task by combining supervised learning and language identification. Experimental results demonstrate that our crawler outperforms previous crawling approaches and produces a high-quality collection of bilingual sites, which we evaluate in the context of machine translation in the tourism and hospitality domain. The parallel text obtained using our novel crawling strategy results in a relative improvement of 22 % in BLEU score (English-to-Spanish) over an out-ofdomain seed translation model trained on the European parliamentary proceedings. 1

In this paper, we present two methods to use a noisy parallel news corpus to improve statistical machine translation (SMT) systems. Taking full advantage of the characteristics of our corpus and of existing resources, we use a bootstrapping strategy, whereby an existing SMT engine is used both to detect parallel sentences in comparable data and to provide an adaptation corpus for translation models. MT experiments demonstrate the benefits of various combinations of these strategies. 1

Abstract. This paper describes an approach to automatically align fragments of texts of two documents in different languages. A text fragment is a list of continuous sentences and an aligned pair of fragments consists of two fragments in two documents, which are content-wise related. Cross-lingual similarity between fragments of texts is estimated based on models of divergence from randomness. A set of aligned fragments based on the similarity scores are selected to provide an alignment between sections of the two documents. Similarity measures based on divergence show strong performance in the context of cross-lingual fragment alignment in the performed experiments.

supervision