Publishing Partner: Cambridge University Press CUP Extra Publisher Login

FYI: Free Ebook on CALL and MT Available


Author: Ziyuan Yao

Linguistic Field(s): Computational Linguistics
Translation

FYI Body: Free ebook available: Breaking the Language Barrier: A Game-
Changing Approach v0.18

Dear All,

I'm pleased to announce my free ebook ''Breaking the Language
Barrier: A Game-Changing Approach'', version 0.18. The ebook has
undergone significant updates since its initial release (version 0.09 in
April 2010). Almost every part has significant new developments, so I
suggest existing readers take it anew.

Perhaps nothing can explain this ebook better than its own Overview,
which is attached below.

View or download it (free) from:

https://sites.google.com/site/yaoziyuan/publications/books/breaking-
the-language-barrier-a-game-changing-approach

Best Regards,

Ziyuan Yao
https://sites.google.com/site/yaoziyuan/

Overview

In today's world, the goal of breaking the language barrier is pursued
on two fronts: language teachers teaching students a second
language, thus enabling humans to manually break the language
barrier, and computational linguists building increasingly better
machine translation systems to automatically break the language
barrier.

However, I see important, unfulfilled opportunities on both fronts:

In second language teaching, amazingly efficient teaching methods
have not gone mainstream and not drawn enough attention from
computational linguists (so that these methods could be automated and
truly powerful). For example, imagine if you're browsing a Web page in
your native language, and a Web browser extension automatically
detects the topic of this page and inserts relevant foreign language
micro-lessons in it, so that you can incidentally learn a foreign
language while browsing interesting native language information :-)
This AdSense-like ''L1-driven L2 teaching'' will be the future of second
language teaching.

In machine translation, computational linguists only pay attention to
computer capabilities to process natural language (known as natural
language processing, NLP), and totally ignore human capabilities to
share some burden from the computer in language processing, which
can lead to significantly better results. For example, theory and
practice have proven that syntax disambiguation is a much harder task
than word sense disambiguation, and therefore machine translation
tends to screw up the word order of the translation result if the
language pair has disparate word orders; but what if machine
translation preserves the source language's word order in the
translation result, and teaches the end user about the source
language's word order so that he can manually figure out the logic of
the translation result? If the end user is willing to commit some of his
own natural intelligence in the man-machine joint effort to break the
language barrier, he will get the job done better.

Therefore this ebook presents emerging ideas and implementations in
computer-assisted language learning (CALL), reading, writing and
machine translation (MT) that strive to leverage both human and
machine language processing potential and capabilities, and will
redefine the way people break the language barrier.

Approaches whose titles have an exclamation mark (!) are stirring
game-changing technologies which are the driving forces behind this
initiative.

Back   FYI main page