22 December 2006

Natural Language Processing

I realized that I haven't blogged about a theoretical topic recently. So for this entry, I decided to blog about something that is related to what I am working on, that is, natural language processing.

First of all, what is NLP?

NLP is a subfield of linguistics and artificial intelligence. It studies how human languages are understood, or processed, by its users, and it tries to model them. Although I do not work on NLP, my research on temporal semantics can be extended to NLP if I ever decide to do so.

So what are some of the research that people doing NLP do?

The research includes pronoun resolution, temporal interpretation, discourse resolution, among other things. But instead of boring my readers with these technical terms, I'd just introduce some of the linguistic puzzles that I have come across the web.

Observe the following two sentences:

  • We gave the monkeys the bananas because they were hungry.

  • We gave the monkeys the bananas because they were over-ripe.

  • In these two sentences, the antecedent of theyare different. In the first sentence, it refers back to the monkeys, but in the second sentence, it refers back to the bananas. The syntax of the two sentences are the same, and the rhetorical relations that exist between the two clauses are also the same (Both are Explanation). But our world knowledge (also known as Knowledge Base) tells us that monkeys cannot be over-ripe and bananas cannot be hungry, thus, it refers back to the proper noun phrase.

  • Time flies like an arrow.

  • This sentence has multiple interpretations. 1) Tiem moves quickly just like an arrow does. 2) Measure the speed of flying insects like you would measure that of an arrow. You should time flies like you would an arrow. 3) Measure the speed of flying insects like an arrow would. Time flies in the same way that an arrow would time them. 4) Measure the speed of flying insects that are like arrows. Time those flies that are like arrows. 5) A type of flying insect "time-fly" enjoys arrows. Fruit flies like a banana.

  • "pretty little girls' school"

  • This one is for you. Who or what is pretty? Who or what is little?

    Finally, let me share the last of my Telephone Series for your enjoyment.

    Enjoy the long weekend!


    1. LIW,
      You musings on NLP problems and subproblems only reminds us what the congregator has said in Ecclesiastes 3:11- that God "has made everything pretty in its time, even time indefinite he has put in our hearts that we may never find out what the true God has made from the start to the finish. So again, you and I can only concur with Solomon's wise conclusion on the matter: that in this vain life, everything is vanity and a striving after wind! -(Ecclesiastes 2:17)

      Ergo,Solomon was ahead of us and everyone else in this NLP human knowledge field. Ought we not remember what his final conclusion was? Ecclesiastes 12:13.

    2. Hi LIW,

      sorry that I didn't check your blog after your last comment on my blog for several days!
      I've been really busy with all preparations for Christmas but now I am nearly ready.
      Your photo is very funny - at he first glance I thought of someone's skin with pimples...


      Hey, Merry Christmas to you!!!!!!!
      I hope you will enjoy the next days and I am looking forward to some photos of course.

    3. These sentences can now be understood via NLP! A Jerusalem-based startup, Linguistic Agents have created a parser that can do it. You can contact the company to view a demo. I played around with it and it actually works.

    4. Hello Pinsan,

      HAPPY NEW YEAR to you! I know you're doing great over there. ^^ May your endeavors be always fruitful. I pray for more blessings to come your way.