Posted on 8th September 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics
CL, Computational Linguistics, ideas, Language Technology
Would it not be a good idea to devise Artificial Intelligence Language systems that rely on the responses from masses of people, rather than trying to generate something meaningful by themselves? Even collecting a few words from hundreds of thousands of people create more intelligence than a system relying on itself to process and give accurate response through different sorts of independent algorithms!
If we are trying to get machines to understand, the question is, do we ourselves understand anyway? Or are there other patterns that make up our brain system, what we call “understanding” while we ourselves don’t understand anyway?
Posted on 8th September 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics
CL, Computational Linguistics, ideas, Language Technology
Perhaps, if a machine can go successfully through finding user-goal in a dialog, the machine could do neat things. The question is, what can the computer do when user goals are broad or too general rather than structured or having a narrow field of ideas
Posted on 8th September 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics
CL, Computational Linguistics, ideas, Language Technology
from:
http://nltk.googlecode.com/svn/trunk/doc/book/ch01.html#ex-set-comprehension-python
At a purely practical level, we all need help to navigate the universe of information locked up in text on the Web. Search engines have been crucial to the growth and popularity of the Web, but have some shortcomings. It takes skill, knowledge, and some luck, to extract answers to such questions as: What tourist sites can I visit between Philadelphia and Pittsburgh on a limited budget? What do experts say about digital SLR cameras? What predictions about the steel market were made by credible commentators in the past week? Getting a computer to answer them automatically involves a range of language processing tasks, including information extraction, inference, and summarization, and would need to be carried out on a scale and with a level of robustness that is still beyond our current capabilities.
On a more philosophical level, a long-standing challenge within artificial intelligence has been to build intelligent machines, and a major part of intelligent behaviour is understanding language. For many years this goal has been seen as too difficult. However, as NLP technologies become more mature, and robust methods for analyzing unrestricted text become more widespread, the prospect of natural language understanding has re-emerged as a plausible goal.
Posted on 26th June 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics
CL, Computational Linguistics, ideas, linguistics, Software, Technology, Translation
It would be good to keep track of linguistics software and to realize how this field can be expanded with new ideas.
A recent link I have come across to is:
and of course, the good old Google search:
Posted on 25th June 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics
CL, Computational Linguistics, ideas
When reading through the exam material, I came to the point where Vinary and Darbelnet’s model of translation was explained (Jeremy Munday: 2001: 56-59) Interestingly enough Vinary and Darbelnet list five steps for the translator to follow in moving from Source Text to Target Text, their first item being “identifying the units of translation”. It is next explained that “as far as the ‘unit of translation’ is concerned, the authors reject the individual word. They consider the unit of translation to be a combination of a lexicological unit and a unit of thought and define it (1995: 21) as: ‘the smallest segment of the utterance whose signs are linked in such a way that they should not be translated individually.”
Now to me, what the authors are suggesting is translation of “smallest units of ideas.” Again, to me this seems very important that a computer program would be able to deal with ideas and recognize smallest units of ideas in a text in order to provide better translations.
Posted on 18th June 2010 by Bamshad (Bob) Lotfabadi in Computational Linguistics |Writings
CL, Computational Linguistics, Google, ideas, Translation
There was a period when I thought Google Translator is the most comprehensive online translation service compared to other similar online systems. Although this is true when comparing it with other “online services”, I am lucky to realize it lacks listing a whole group of words when it comes to translating from English to Farsi. I’m taking a “morphology” course this semester where the students are requested to find Persian meanings to English words containing Greek and Latin roots, prefixes, and suffixes.
When trying to utilize Google Translator to find meanings for this whole list of new words that pop-up throughout the course, I see roughly-speaking that over 70 percent of the words are not defined in Google Translator. The translator could be improved a lot and to me, this is an important conclusion!
