Unsupervised Word Mapping Using Structural Similarities in Monolingual Embeddings
share
Most existing methods of automatic bilingual dictionary induction rely on prior alignments between the source and target languages, such as parallel corpora or seed dictionaries. For many language pairs, such supervised alignments are not readily available. We propose an unsupervised approach for learning a bilingual dictionary for a pair of languages given their independently-learned monolingual word embeddings. The proposed method exploits local and global structures in monolingual vector spaces to align them such that similar words are mapped to each other. We show experimentally that the performance of the bilingual alignments learned using the unsupervised method is comparable to supervised bilingual alignments using a seed dictionary.
READ FULL TEXT