How much hybridization does machine translation Need?

Abstract

Rule-based and corpus-based machine translation (MT) have coexisted for more than 20 years. Recently, boundaries between the two paradigms have narrowed and hybrid approaches are gaining interest from both academia and businesses. However, since hybrid approaches involve the multidisciplinary interaction of linguists, computer scientists, engineers, and information specialists, understandably a number of issues exist.While statistical methods currently dominate research work in MT, most commercial MT systems are technically hybrid systems. The research community should investigate the benefits and questions surrounding the hybridization of MT systems more actively. This paper discusses various issues related to hybrid MT including its origins, architectures, achievements, and frustrations experienced in the community. It can be said that both rule-based and corpus-based MT systems have benefited from hybridization when effectively integrated. In fact, many of the current rule/corpus-based MT approaches are already hybridized since they do include statistics/rules at some point.

Machine Translation using Semantic Web Technologies: A Survey

“…• Semantic Web Journal (SWJ) 11 • Journal of Web Semantics (JWS) 12 • Machine Translation Journal (MT) 13 4 The search engines automatically take into account the inflections and synonyms of the keywords.…”

“…This combination of methods is called hybrid MT. Although hybrid approaches have been achieving good results, they still suffer from some RBMT problems [10][11][12], for example, the big effort of adding new rules for handling a given syntax divergence. Nowadays, a novel SMT paradigm has arisen called Neural Machine Translation (NMT) which relies on Neural Network (NN) algorithms.…”

From Feature to Paradigm: Deep Learning in Machine Translation (Extended Abstract)

“…In rule-based MT, limitations are that it requires many linguistic resources, and a lot of human expert time. There is a considerable amount of research trying to hybridize these two approaches [Costa-jussà, 2015].…”