Statistical Machine Translation between Related Languages

Language­independent Statistical Machine Translation (SMT) has proven to be very challenging. The diversity of languages makes high accuracy difficult and requires substantial parallel corpus as well as linguistic resources (parsers, morph analyzers, etc.). An interesting observation is that a large chunk of machine translation (MT) requirements involve related languages. They are either : (i) between related languages, or (ii) between a lingua franca (like English) and a set of related languages. For instance, India, the European Union and South­East Asia have such translation requirements due to government, business and socio­cultural communication needs. Related languages share a lot of linguistic features and the divergences among them are at a lower level of the NLP pipeline. The objective of the tutorial is to discuss how the relatedness among languages can be leveraged to bridge this language divergence thereby achieving some/all of these goals: (i) improving translation quality, (ii) achieving better generalization, (iii) sharing linguistic resources, and (iv) reducing resource requirements. We will look at the existing research in SMT from the perspective of related languages, with the goal to build a toolbox of methods that are useful for translation between related languages. This tutorial would be relevant to Machine Translation researchers and developers, especially those interested in translation between low­resource languages which have resource­rich related languages. It will also be relevant for researchers interested in multilingual computation. We start with a motivation for looking at the SMT problem from the perspective of related languages. We introduce notions of language relatedness useful for MT. We explore how lexical, morphological and syntactic similarity among related languages can help MT. Lexical similarity will receive special attention since related languages share a significant vocabulary in terms of cognates, loanwords, etc. Then, we look beyond bilingual MT and present how pivot­based and multi­source methods incorporate knowledge from multiple languages, and handle language pairs lacking parallel corpora. We present some studies concerning the implications of languages relatedness to pivot­based SMT, and ways of handling language divergence in the pivot­based SMT scenario. Recent advances in deep learning have made it possible to train multi­language neural MT systems, which we think would be relevant to training between related languages.