What is the logic behind a web language tranlatior? just word maching?

I would like to create a simple language translator program, English and Japanese.

Is the logic behind current web translators just simple a word-by-word matching?

Are there any other or better programming logic I can use to create a translator?

Do you think this is the only way we can create a language translator no matter how our technogy advances in the future?What is the logic behind a web language tranlatior? just word maching?
Hi,



The best language translators I have ever encountered for web content have been human, manual translations.



Of course, this isn't optimal. So, the second best I have used is Google's, actually fairly extensive, API for translating. It was good with certain languages, like Spanish, but when it came to others, like Russian, it couldn't match human translation.



Hope this helps!



- ChaseWhat is the logic behind a web language tranlatior? just word maching?
I am sure it is more complex than just word matching, e.g. context sensitive to disambiguate homonyms.

With better technology, the translator can spend more time searching databases for usage examples etc.What is the logic behind a web language tranlatior? just word maching?
there are several ways to go about a language translator. you can use compiler technology using a lexical analyzer and a parser with syntax-directed translation. Or you could use Artificial Intelligence or Fuzzy Logic or Neural Networks or one of the myriad other ways of handling this sort of task.



frankly, it's been done and I don't know what IBM used. IBM's Watson supercomputer recognizes natural spoken language and can also respond in like manner. Albeit with a Hal voice.