tailieunhanh - Báo cáo khoa học: "An Alignment Algorithm using Belief Propagation and a Structure-Based Distortion Model"

In this paper, we first demonstrate the interest of the Loopy Belief Propagation algorithm to train and use a simple alignment model where the expected marginal values needed for an efficient EM-training are not easily computable. We then improve this model with a distortion model based on structure conservation. | An Alignment Algorithm using Belief Propagation and a Structure-Based Distortion Model Fabien Cromieres Graduate school of informatics Kyoto University Kyoto Japan fabien@ Sadao Kurohashi Graduate school of informatics Kyoto University Kyoto Japan kuro@ Abstract In this paper we first demonstrate the interest of the Loopy Belief Propagation algorithm to train and use a simple alignment model where the expected marginal values needed for an efficient EM-training are not easily computable. We then improve this model with a distortion model based on structure conservation. 1 Introduction and Related Work Automatic word alignment of parallel corpora is an important step for data-oriented Machine translation whether Statistical or Example-Based as well as for automatic lexicon acquisition. Many algorithms have been proposed in the last twenty years to tackle this problem. One of the most successfull alignment procedure so far seems to be the so-called IBM model 4 described in Brown et al. 1993 . It involves a very complex distortion model here and in subsequent usages distortion will be a generic term for the reordering of the words occurring in the translation process with many parameters that make it very complex to train. By contrast the first alignment model we are going to propose is fairly simple. But this simplicity will allow us to try and experiment different ideas for making a better use of the sentence structures in the alignment process. This model and even more so its subsequents variations although simple do not have a computationally efficient procedure for an exact EM-based training. However we will give some theoretical and empirical evidences that Loopy Belief Propagation can give us a good approximation procedure. Although we do not have the space to review the many alignment systems that have already been proposed we will shortly refer to works that share some similarities with our approach. In particular the .