- Abstract
- An interactive framework that integrates the human knowledge into the translation process has been presented in previous work.
- We describe a new interactive machine translation approach that is able to work with phrase-based and hirerchical traslation models, and integrates error-corection all in a unified statistical framework.
- Introduction
- Two basic problems exist:
- The first problem arises when the system cannot generate the prefix validated by the user.(propose an alternative error-correction formalization and a specific error-correction model).
- The second problem is how the system deals with word reordering. (we introduce the use of hierarchical MT models).
- Two basic problems exist:
- Statistical Framework
- Statistical Machine Translation
- Statistical Interactive Machine Translation
- The crucial step of the process is the production of the suffix.
which can be straighforwardly rewritten as:
- IMT with Stochastic Error-Collection
we finally arrive to the expression:
![]()
- Alternative Formalization for IMT with Stochastic Error-Correction