Re: Unicode & space in programming & l10n

From: John D. Burger (john@mitre.org)
Date: Wed Sep 27 2006 - 09:14:39 CST

  • Next message: Otto Stolz: "Re: Problem with SSI and BOM"

    Jefsey_Morfin wrote:

    > Another problem is that English is a
    > complex language to automatically understand. This increases the cost
    > of the English globalization or reduce its possibilities.

    After twenty years of working on natural language processing, I am
    familiar with no research that shows that English is inherently more
    complex to process than any other language. In fact, in modern
    approaches, the single biggest factor is the amount of data available
    for training. This is especially so for statistical machine
    translation. This often makes English easier to process than other
    languages, due simply to the overwhelming availability of training
    material.

    - John D. Burger
       MITRE



    This archive was generated by hypermail 2.1.5 : Wed Sep 27 2006 - 09:17:20 CST