Dynamical visualization of nested correspondences

Introduction Computer-aided translation systems based on translation memories have spread rapidly in the translation industry over the last 15 years or so. A key reason for their success is that they require no language-specific resources. One of their main drawbacks is that they do not allow translators to reuse translation of micro-examples that are abundant in the memories. Indeed, in the classical use of translation memories, a fuzzy match will only be presented to the translator if it broadly corresponds to the segment to be translated. This granularity is far from optimal to take advantage of minor correspondences that will remain ignored by the system. These issues have been repeatedly addressed in the literature; see for example [7], [13], [20], [25], [26] and [27].