Cross Sentence Alignment for Structurally Dissimilar Corpus Based on Singular Value Decomposition

Extracting the alignment pairs is a critical step for constructing bilingual corpus knowledge base for Example Based Machine Translation Systems. Different methods have been proposed in aligning parallel corpus between two different languages. However, most of them focus on structurally similar languages like English-French. This paper presents a method of cross aligning Portuguese-Chinese bilingual comparable corpus. The proposed approach is based on Singular Value Decomposition techniques and similarity measurement, which covers the problem in aligning structurally dissimilar corpus and enhances the accuracy of the alignment result.