Using machine translation output as a starting point for human translation has recently gained traction in the transla-tion community. This paper describes cdec Realtime, a framework for build-ing adaptive MT systems that learn from post-editor feedback, and TransCenter, a web-based translation

Schema matching is a critical step in many applications, such as XML message mapping, data warehouse loading, and schema integration. In this paper, we investigate algorithms for generic schema matching, outside of any particular data model or application. We first present a taxonomy for past solutions, showing that a rich range of techniques is available. We then propose a new algorithm, Cupid, that discovers mappings between schema elements based on their names, data types, constraints, and schema structure, using a broader set of techniques than past approaches. Some of our innovations are the integrated use of linguistic and structural matching, context-dependent matching of shared types, and a bias toward leaf structure where much of the schema content resides. After describing our algorithm, we present experimental results that compare Cupid to two other schema matching systems.

to improve the speed of human translation has gained interest. In the frequently employed practice of post-editing, a machine translation system outputs an initial translation and a human translator edits it for correctness, ideally saving time over translating from scratch. While general improvements in MT

In this thesis we investigate the hierarchical phrase-based approach to machine translation, with special attention to the search problem. This approach is nowadays one of the most widely applied for statistical machine translation, and thus a detailed study helps in advancing the state

the task is to find material relevant to a given information need. Cross-Language Information Retrieval (CLIR) is a special case of IR when the search takes place in a multi-lingual collection. Of course, it is not helpful to retrieve content in languages the user cannot understand. Machine Translation (MT

During my time as a PhD student at the University of Minnesota, I often told others that I was the only full-time machine researcher between Pittsburgh and Stanford. While I eventually learned of a handful of exceptions to that statement (most notably in Dayton, Ohio, where I am now employed

This document serves as user manual and code guide for the Moses machine translation decoder. The decoder was mainly developed by Hieu Hoang and Philipp Koehn at the University of Edinburgh and extended during a Johns Hopkins University Summer Workshop and further developed under Euro

second edition

of translation, reprinting, re-use of illustrations, broadcasting, reproduction by photocopying machines or similar means, and storage in data banks. Copyright c ○ 2010 by the authors

This paper presents a new discriminative model for information retrieval (IR), referred to as linear discriminant model (LDM), which provides a flexible framework to incorporate arbitrary features. LDM is different from most existing models in that it takes into account a variety of linguistic features that are derived from the component models of HMM that is widely used in language modeling approaches to IR. Therefore, LDM is a means of melding discriminative and generative models for IR. We present two algorithms of parameter learning for LDM. One is to optimize the average precision (AP) directly using an iterative procedure. The other is a perceptron-based algorithm that minimizes the number of discordant document-pairs in a rank list. The effectiveness of our approach has been evaluated on the task of ad hoc retrieval using six English and Chinese TREC test sets. Results show that (1) in most test sets, LDM significantly outperforms the state-of-the-art language modeling approaches and the classical probabilistic retrieval model; (2) it is more appropriate to train LDM using a measure of AP rather than likelihood if the IR system is graded on AP; and (3) linguistic features (e.g. phrases and dependences) are effective for IR if they are incorporated properly.