Bitext alignment for the English-Ukrainian language pair
Zoryana Paliy, Andriy Romanyuk · International Conference on Experience of Designing and Applications of CAD Systems in Microelectronics · 2009
In this paper, the problem of bitext alignment as the main stage of a parallel corpus composition is addressed. Three automatic alignment algorithms are chosen for evaluation in their program implementations. Bitexts for the alignment are composed of a total of 736 sentence pairs (28,303 words) from English fiction texts and their corresponding translations. Automatic alignment of the chosen bitexts is performed and evaluated. The results of the aligners' performance are summarized and discussed.