Triangulation of Reordering Tables: An Advancement Over Phrase Table Triangulation in Pivot-Based SMT
Deepak U. Patil, Harshad Chavan, Pushpak Bhattacharyya · 2015
Triangulation in Pivot-Based Statistical Machine Translation(SMT) is a very effec-tive method for building Machine Trans-lation(MT) systems in case of scarcity of the parallel corpus. Phrase Table Triangu-lation helps in such a resource constrained setting by inducing new phrase pairs with the help of a pivot. However, it does not explore the possibility of extracting re-ordering information through the use of pivot. This paper presents a novel method for triangulation of reordering tables in Pivot Based SMT. We show that the use of a pivot can help in extracting better re-ordering information and can assist in im-proving the quality of the translation. With a detailed example, we show that triangu-lation of reordering tables also improves the lexical choices a system makes during translation. We observe a BLEU score im-provement of 1.06 for Marathi to English MT system with Hindi as a pivot, and also significant improvements in 8 other trans-lation systems by using this method. 1