Most research in Machine translation is about having the computers completely bear the load of translating one human language into another. This paper looks at the machine translation problem afresh and observes that there is a need to share the load between man and machine, distinguish ‘reliable’ knowledge from the ‘heuristics’, provide a spectrum of outputs to serve different strata of people, and finally make use of existing resources instead of reinventing the wheel. This paper describes the architecture and design based on the fundamental premise of sharing the load, resulting in “good enough” results according to the needs of the reader. The architecture differs from the conventional in three major ways: 1. Reversal in the order of operations as compared to conventional machine translation systems 2. Introduction of interfaces that act like glue and improve the modularity of the system 3. Development of a GUI to provide the ‘right ‘ amount of information at the right time The paper attempts to prove that this new architecture is a better approach to Machine translation process transparent to the user-cum-developer; and it leads to machine translation in stages, thus ensuring robustness.
Machine translation, sometimes referred to by the abbreviation MT, is a sub-field of computational linguistics that investigates the use of computer software to translate text or speech from one natural language to another. At its basic level, MT performs simple substitution of words in one natural language for words in another. Using corpus techniques, more complex translations may be attempted, allowing for better handling of differences in linguistic typology, phrase recognition, and translation of idioms, as well as the isolation of anomalies. Current machine translation software often allows for customization by domain or profession (such as weather reports) — improving output by limiting the scope of allowable substitutions. This technique is particularly effective in domains where formal or formulaic language is used. It follows then that machine translation of government and legal documents more readily produces usable output than conversation or less standardized text. Improved output quality can also be achieved by human intervention: for example, some systems are able to translate more accurately if the user has unambiguously identified which words in the text are names. With the assistance of these techniques, MT has proven useful as a tool to assist human translators and, in a very limited number of cases, can even produce output that can be used as is (e.g., weather reports). The idea of machine translation may be traced back to the 17th century. In 1629, René Descartes proposed a universal language, with equivalent ideas in different tongues sharing one symbol. In the 1950s, The Georgetown experiment (1954) involved fully-automatic translation of over sixty Russian sentences into English. The experiment was a great success and ushered in an era of substantial funding for machine-translation research. The authors claimed that within three to five years, machine translation would be a solved problem. Real progress was much slower, however, and after the ALPAC report (1966), which found that the ten-year-long research had failed to fulfill expectations, funding was greatly reduced. Beginning in the late 1980s, as computational power increased and became less expensive, more interest was shown in statistical models for machine translation. The idea of using digital computers for translation of natural languages was proposed as early as 1946 by A. D. Booth and possibly others. The Georgetown experiment was by no means the first such application, and a demonstration was made in 1954 on the APEXC machine at Bareback College (University of London) of a rudimentary translation of English into French. Several papers on the topic were published at the time, and even articles in popular...
Please join StudyMode to read the full document