We present a language agnostic, unsupervised method for inducing morphological transformations between words.The method relies on certain regularities manifest in highdimensional vector spaces.We show that this method is capable of discovering a wide range of morphological rules, which in turn are used to build morphological analyzers.We evaluate this method across six different languages and nine datasets, and show significant improvements across all languages.

In this paper we describe two new objective automatic evaluation methods for machine translation. The first method is based on longest common subsequence between a candidate translation and a set of reference translations. Longest common subsequence takes into account sentence level structure similarity naturally and identifies longest co-occurring in-sequence n-grams automatically. The second method relaxes strict n-gram matching to skip-bigram matching. Skip-bigram is any pair of words in their sentence order. Skip-bigram cooccurrence statistics measure the overlap of skip-bigrams between a candidate translation and a set of reference translations. The empirical results show that both methods correlate with human judgments very well in both adequacy and fluency.

A phrase-based statistical machine translation approach — the alignment template approach — is described. This translation approach allows for general many-to-many relations between words. Thereby, the context of words is taken into account in the translation model, and local changes in word order from source to target language can be learned explicitly. The model is described using a log-linear modeling approach, which is a generalization of the often used source-channel approach. Thereby, the model is easier to extend than classical statistical machine translation systems. We describe in detail the process for learning phrasal translations, the feature functions used, and the search algorithm. The evaluation of this approach is performed on three different tasks. For the German-English speech Verbmobil task, we analyze the effect of various system components. On the French-English Canadian Hansards task, the alignment template system obtains significantly better results than a single-word-based translation model. In the Chinese-English 2002 National Institute of Standards and Technology (NIST) machine translation evaluation it yields statistically significantly better NIST scores than all competing research and commercial translation systems.

Often, the training procedure for statistical machine translation models is based on maximum likelihood or related criteria. A general problem of this approach is that there is only a loose relation to the final translation quality on unseen text. In this paper, we analyze various training criteria which directly optimize translation quality. These training criteria make use of recently proposed automatic evaluation metrics. We describe a new algorithm for efficient training an unsmoothed error count. We show that significantly better results can often be obtained if the final evaluation criterion is taken directly into account as part of the training procedure.

Statistical techniques for machine translation offer promise for rapid development in response to unexpected requirements, but realizing that potential requires rapid acquisition of required resources as well. This paper reports the results of experiments with resources collected in ten days; about 1.3 million words of parallel text from five types of sources and a bilingual term list with about 20,000 term pairs. Systems were trained with resources individually and in combination, using an approach based on alignment templates. The use of all available resources was found to yield the best results in an automatic evaluation using the BLEU measure, but a single resource (the Bible) coupled with a small amount of in-domain manual translation (less than 6,000 words) achieved more than 85% of that upper baseline. With a concerted effort, such a system could be built in a single day.

We present and compare various methods for computing word alignments using statistical or heuristic models. We consider the five alignment models presented in Brown, Della Pietra, Della Pietra, and Mercer (1993), the hidden Markov alignment model, smoothing techniques, and refinements. These statistical models are compared with two heuristic models based on the Dice coefficient. We present different methods for combining word alignments to perform a symmetrization of directed statistical alignment models. As evaluation criterion, we use the quality of the resulting Viterbi alignment compared to a manually produced reference alignment. We evaluate the models on the German-English Verbmobil task and the French-English Hansards task. We perform a detailed analysis of various design decisions of our statistical alignment system and evaluate these on training corpora of various sizes. An important result is that refined alignment models with a first-order dependence and a fertility model yield significantly better results than simple heuristic models. In the Appendix, we present an efficient training algorithm for the alignment models presented.

We present C*ST*RD, a cross-language information delivery system that supports cross-language information retrieval, information space visualization and navigation, machine translation, and text summarization of single documents and clusters of documents. C*ST*RD was assembled and trained within 1 month, in the context of DARPA's Surprise Language Exercise, that selected as source a heretofore unstudied language, Hindi. Given the brief time, we could not create deep Hindi capabilities for all the modules, but instead experimented with combining shallow Hindi capabilities, or even English-only modules, into one integrated system. Various possible configurations, with different tradeoffs in processing speed and ease of use, enable the rapid deployment of C*ST*RD to new languages under various conditions.

We present a framework for statistical machine translation of natural languages based on direct maximum entropy models, which contains the widely used source-channel approach as a special case. All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible hidden variables. This approach allows a baseline machine translation system to be extended easily by adding new feature functions. We show that a baseline statistical machine translation system is significantly improved using this approach.

Statistical machine translation systems usually compute the single sentence that has the highest probability according to the models that are trained on data. We describe a method for constructing a word graph to represent alternative hypotheses in an efficient way. The advantage is that these hypotheses can be rescored using a refined language or translation model. Results are presented on the German-English Verbmobil corpus.

Typically, the lexicon models used in statistical machine translation systems do not include any kind of linguistic or contextual information, which often leads to problems in performing a correct word sense disambiguation. One way to deal with this problem within the statistical framework is to use maximum entropy methods. In this paper, we present how to use this type of information within a statistical machine translation system. We show that it is possible to significantly decrease training and test corpus perplexity of the translation models. In addition, we perform a rescoring of N-Best lists using our maximum entropy model and thereby yield an improvement in translation quality. Experimental results are presented on the so-called "Verbmobil Task".

Over the past years, automatic dialogue systems and telephonebased machine inquiry systems have received increasing attention. In addition to an automatic speech recognizer and a dialogue manager, such systems consist of a natural language understanding (NLU) component. Some of the most investigated approaches to NLU are rule-based methods as Stochastic Grammars, which are often written manually. However, the sole usage of rule-based methods can turn out to be inflexible and the problem of reusability occurs. When extending the application scenario or changing the application's domain itself, a large part of the set of rules often must be rewritten. Therefore, techniques are desirable which help to reduce the manual effort when building up an NLU component for a new domain. In this paper we investigate an approach to NLU, which is derived from the field of statistical machine translation. Starting from a conceptual annotated corpus, we describe the problem of NLU as a translation from a source sentence to a formallanguage target sentence. Doing this, we will mainly focus on the quality of different alignment models between source and target sentences. Even though the usage of grammars cannot be totally avoided in NLU-systems, it is our goal to reduce their employment and learn the dependencies between words and their meaning automatically. Experiments were performed on the Philips in-house TABA corpus, which is a text corpus in the domain of a German train timetable information system. 1.

We describe methods for translating a text given in multiple source languages into a single target language. The goal is to improve translation quality in applications where the ultimate goal is to translate the same document into many languages. We describe a statistical approach and two specific statistical models to deal with this problem. Our method is generally applicable as it is independent of specific models, languages or application domains. We evaluate the approach on a multilingual corpus covering all eleven official European Union languages that was collected automatically from the Internet. In various tests we show that these methods can significantly improve translation quality. As a side effect, we also compare the quality of statistical machine translation systems for many European languages in the same domain.

In this paper, we present and compare various single-word based alignment models for statistical machine translation. We discuss the five IBM alignment models, the Hidden-Markov alignment model, smoothing techniques and various modifications. We present different methods to combine alignments. As evaluation criterion we use the quality of the resulting Viterbi alignment compared to a manually produced reference alignment. We show that models with a first-order dependence and a fertility model lead to significantly better results than the simple models IBM-1 or IBM-2, which are not able to go beyond zero-order dependencies.

We describe three approaches to statistical translation and present experimental results. The statistical translation approach uses two types of information: a translation model and a language model. The language model used is a bigram or general m-gram model. The translation model is decomposed into a lexical model and an alignment model. There are three approaches that are presented and tested in detail: the quasi-monotone alignment approach, the inverted alignment approach, and the alignment template approach. For each of these three approaches, a suitable search method is presented. The system has been tested on a limited-domain spoken-language task for which a bilingual corpus is available: the Verbmobil task (German-English, 7000-word vocabulary). We present experimental results for each of the three approaches. The experimental tests were performed on both the text transcription and the speech recognizer output.