Major issues machine translation
Word-sense disambiguation concerns finding a suitable translation when a word can have more than one meaning. The problem was first raised in the 1950s by Yehoshua Bar-Hillel. He pointed out that without a "universal encyclopedia", a machine would never be able to distinguish between the two meanings of a word. Today there are numerous approaches designed to overcome this problem. They can be approximately divided into "shallow" approaches and "deep" approaches.
Shallow approaches assume no knowledge of the text. They simply apply statistical methods to the words surrounding the ambiguous word. Deep approaches presume a comprehensive knowledge of the word. So far, shallow approaches have been more successful.
Claude Piron, a long-time translator for the United Nations and the World Health Organization, wrote that machine translation, at its best, automates the easier part of a translator's job; the harder and more time-consuming part usually involves doing extensive research to resolve ambiguities in the source text, which the grammatical and lexical exigencies of the target language require to be resolved:
Why does a translator need a whole workday to translate five pages, and not an hour or two? ..... About 90% of an average text corresponds to these simple conditions. But unfortunately, there's the other 10%. It's that part that requires six [more] hours of work. There are ambiguities one has to resolve. For instance, the author of the source text, an Australian physician, cited the example of an epidemic which was declared during World War II in a "Japanese prisoner of war camp". Was he talking about an American camp with Japanese prisoners or a Japanese camp with American prisoners? The English has two senses. It's necessary therefore to do research, maybe to the extent of a phone call to Australia.
The ideal deep approach would require the translation software to do all the research necessary for this kind of disambiguation on its own; but this would require a higher degree of AI than has yet been attained. A shallow approach which simply guessed at the sense of the ambiguous English phrase that Piron mentions (based, perhaps, on which kind of prisoner-of-war camp is more often mentioned in a given corpus) would have a reasonable chance of guessing wrong fairly often. A shallow approach that involves "ask the user about each ambiguity" would, by Piron's estimate, only automate about 25% of a professional translator's job, leaving the harder 75% still to be done by a human.
One of the major pitfalls of MT is its inability to translate non-standard language with the same accuracy as standard language. Heuristic or statistical based MT takes input from various sources in standard form of a language. Rule-based translation, by nature, does not include common non-standard usages. This causes errors in translation from a vernacular source or into colloquial language. Limitations on translation from casual speech present issues in the use of machine translation in mobile devices.
Name entities, in narrow sense, refer to concrete or abstract entities in the real world including people, organizations, companies, places etc. It also refers to expressing of time, space, quantity such as 1 July 2011, $79.99 and so on.
Named entities occur in the text being analyzed in statistical machine translation. The initial difficulty that arises in dealing with named entities is simply identifying them in the text. Consider the list of names common in a particular language to illustrate this – the most common names are different for each language and also are constantly changing. If named entities cannot be recognized by the machine translator, they may be erroneously translated as common nouns, which would most likely not affect the BLEU rating of the translation but would change the text's human readability.It is also possible that, when not identified, named entities will be omitted from the output translation, which would also have implications for the text's readability and message.
Another way to deal with named entities is to use transliteration instead of translation, meaning that you find the letters in the target language that most closely correspond to the name in the source language. There have been attempts to incorporate this into machine translation by adding a transliteration step into the translation procedure. However, these attempts still have their problems and have even been cited as worsening the quality of translation. Named entities were still identified incorrectly, with words not being transliterated when they should or being transliterated when they shouldn't. For example, for "Southern California" the first word should be translated directly, while the second word should be transliterated. However, machines would often transliterate both because they treated them as one entity. Words like these are hard for machine translators, even those with a transliteration component, to process.
The lack of attention to the issue of named entity translation has been recognized as potentially stemming from a lack of resources to devote to the task in addition to the complexity of creating a good system for named entity translation. One approach to named entity translation has been to transliterate, and not translate, those words. A second is to create a "do-not-translate" list, which has the same end goal – transliteration as opposed to translation. Both of these approaches still rely on the correct identification of named entities, however.
A third approach to successful named entity translation is a class-based model. In this method, named entities are replaced with a token to represent the class they belong to. For example, "Ted" and "Erica" would both be replaced with "person" class token. In this way the statistical distribution and use of person names in general can be analyzed instead of looking at the distributions of "Ted" and "Erica" individually. A problem that the class based model solves is that the probability of a given name in a specific language will not affect the assigned probability of a translation. A study by Stanford on improving this area of translation gives the examples that different probabilities will be assigned to "David is going for a walk" and "Ankit is going for a walk" for English as a target language due to the different number of occurrences for each name in the training data. A frustrating outcome of the same study by Stanford (and other attempts to improve named recognition translation) is that many times, a decrease in the BLEU scores for translation will result from the inclusion of methods for named entity translation.