Similarities and Differences 论文

2005引用 346
Natural Language Processing TechniquesMachine Learning and AlgorithmsTopic Modeling

摘要

A match sequence between two sentences of a language consists of similarities and di erences to represent similar parts and differing parts between those sentences. A translation example is a corresponding of two sentences in which one of them from the language L a and the other one from the language L b. From given two translation examples we create a match sequence for sentences of the language L a, and another match sequence for sentences of the language L b. From these match sequences, general templates are learned using certain learning heuristics. These heuristics replace di erences or similarities in the match sequences, and establish bindings between variables to create a general template. These learned translation templates can be used in the translation of other sentences in both directions.

作者

暂无数据

相关事件

暂无数据

相关文章

暂无数据