Word reordering as a preprocessor for machine translation systems
R. N. Devendrakumar, A. Praveena · 2017 International Conference on Inventive Systems and Control (ICISC) · 2017
Existing machine translation systems work well when the structures of both the source and target languages are similar. Generally, Indian languages follow Subject-Object-Verbstructure. When translating English sentences to Indian languages, we propose that reordering the sentences to SOV as a preprocessing step will improve the performance of the translation.