A verbal thought (Literally)

Wrote this in my company blog today and since this space keeps languishing every once in a while, I thought I'll update it here

How does the human brain process speech? How does it understand the semantics of language? If there are ambiguities, what mechanisms does it use to resolve? For example lets have a look at these sentences, He is a man of simple wants and He wants money.

Admitted this might be more for people who are interested in Artificial Intelligence or probably wonder how to build systems that understand text. In one of the scenarios that I was working, I came across something fascinating. The words need and want are very interesting. As a noun they really don't require an object action to finish the sentence. They themselves become synonyms. For example look at the sentence He is a man of simple needs. Assuming a machine can process that needs is almost a noun here and can identify the underlying meaning behind needs, you have an absolutely great semantic engine.(I'd use the word semantic as it makes more sense :-) ).

Now here are two interesting sentences :

I need a camera.
I want a car.

In both the cases the verb is a simple verb and requires an object i.e camera or car to complete the sentence. Processing them is a breeze for some of the systems today. Heck! even a rudimentary system developed by us can process these kinds of sentences.

Lets get into a more specific scenario. Now the goal is to understand how verbs like need and want behave. Lets introduce a bit of anarchy (literally) and see if these words behave just the same?.

I need to purchase a car
I want to buy a house in Chennai.

Notice anything different? At first glance you see that there are two new verbs introduced purhcase and buy. Well both of them essentially means the same. You could also add construct to the second sentence. Now you have two verbs in the same sentence. How does your brain resolve? Do these steps happen?

1) Decide that need takes in an object to complete it's meaning.
2) Looking for an object action, come to purchase a car, buy a house.
3) Realize that purchase and buy are two verbs which in turn require an object action to complete their sentence. One cannot literally say I want to buy. There is always the question of What action that follows the verb.
4) Then decide that want to purchase actually points to a desire to verb. i.e desire to purchase.

Essentially when faced with words like want, need, a system cannot just process them alone. It has to look for the other verbs in the sentence which aid in their meaning. So the secondary verb is of more importance or is the deciding factor in the sentence. If any of you can help me figuring out much complex situations, you are welcome. I just am pouring out my thoughts here.

One should marvel at the wonders of the human brain. All of us could handle these sentences even in our first standard. In trying to mimic the brain, I just am stumped at how the brain handles complex things in such simpler steps. As for me, I think finishing the Wren and Martin book again, with a deeper understanding of each of the parts of speech is something that cannot be put down for ever :-).

Correction:I had written this in my company internal blog and Rb corrected me at the parts where I have crossed the text. This is his comment :

In my view it is "I need to act" and that action could be "run" or "purchase a house". The What is not mandatory. The "house" to me belongs to the action of "purchase" and not of "need".

Though "I want to purchase a house" is semantically similar to "I want a house"; they cannot be treated the same. Examples of the form "I went to purchase a house" or "I want to sell my house" show the tie-in of the "house" to the "action" and away from the "want".

The varied streams of reasoning

A daily quiz question mail-list sent a question with three pictures, asking us to give a connect between the three. I found that the first two pictures just gave away the connect. The first was a picture of Stephen King, second was a picture of Rita Hayworth. To anybody who has watched The Shawshank redemption, this would be a piece of cake. I guessed the third picture to be Frank Darabont, which was probably inferred from the fact, that I know Stephen King and Rita Hayworth connect only for this movie and the other closest connect I can find is Frank Darabont. It turned out to be right when I verified my thoughts.

What exactly triggered my brain to give the answer as Frank Darabont. At a mathematical level I'd say that the intersection of {King,Hayworth,The Shawshank Redemption} can lead me to certain connects like {Morgan Freeman,Bob Gunton, Frank Darabont,...}. Now I know that the third picture is not Bob Gunton or Morgan Freeman because I know how they look like. So the only possible connect here was Darabont. While we can mathematically eliminate a set of choices, the visual recognition also plays a huge part in the final elimination here. Yesterday's question was similar. I had to look at an epitaph of a young poet to figure out whom the epitaph belonged to. The young poets I knew, who died early in life were {Keats,Wilfred Owen,...}. I decided to look at the date of his death. Now that leads me to only Victorian Era poets. Here the only Victorian Era poet who died and closely resembled the epitaph was Keats. That turned out to be the answer. What amazes me is the fact that so much stimuli are hit upon when our brain is asked to process or deduce an answer. The cognitive processes that follow deal with visual recognition, context understanding, elimination of other choices.

These are minimal observations from somebody who also is a budding practitioner of Artificial Intelligence, but the questions these observations throw are mind boggling. How do I for example build a simple reasoning engine which can also accumulate knowledge. From the previous examples the accumulating knowledge part was how Frank Darabont and Keats get added to the brain. Now they are stored permanently in my brain and probably would serve for future knowledge gathering.

As Steven Pinker put it, The main lesson of thirty-five years of AI research is that the hard problems are easy and the easy problems are hard. The mental abilities of a four-year-old that we take for granted – recognizing a face, lifting a pencil, walking across a room, answering a question – in fact solve some of the hardest engineering problems ever conceived.... As the new generation of intelligent devices appears, it will be the stock analysts and petrochemical engineers and parole board members who are in danger of being replaced by machines. The gardeners, receptionists, and cooks are secure in their jobs for decades to come

The analogy of a Human Brain

No great content ahead folks. So I just did a Brain.rand(MyBrain) to come up with something.What turned out were highly unusual. I thought my brain was just another empty shell but it's seemingly decent storage capacity did return a myriad of content. So here goes a few random thoughts.

I was leafing through A Study in Scarlet by Arthur Conan Doyle where for the first time he introduces Dr.John.H Watson and Sherlock Holmes, the greatest fictional detective ever. Watson is piqued at the fact that Holmes knows nothing about the earth or the sun or even anything about Poetry and Carlyle. Did a google for the exact text and I got it...

"His ignorance was as remarkable as his knowledge. Of contemporary literature, philosophy and politics he appeared to know next to nothing. Upon my quoting Thomas Carlyle, he inquired in the naivest way who he might be and what he had done. My surprise reached a climax, however, when I found incidentally that he was ignorant of the Copernican Theory and of the composition of the Solar System. That any civilized human being in this nineteenth century should not be aware that the earth"


This really set me on the contemplative mood. How can Sherlock Holmes to whom knowledge and perfection is of paramount importance can not be aware of this basic fact? Why was he not too keen to update his knowledge?So many questions just raced through my mind with each one trying to win the race. The next few paragraphs cleared my questions. Surprisingly enough they made me look at learning in a different perspective. Just read the following extracts from the novel and you'll get a wind of it.

Sherlock Holmes : "You appear to be astonished," he said, smiling at my expression of surprise. "Now that I do know it I shall do my best to forget it."

Watson: "To forget it!"

Sherlock Holmes : "You see," he explained, I consider that a man's brain originally is like a little empty attic, and you have to stock it with such furniture as you choose. A fool takes in all the lumber of every sort that he comes across, so that the knowledge which might be useful to him gets crowded out, or at best is jumbled up with a lot of other things, so that he has a difficulty in laying his hands upon it. Now the skillful workman is very careful indeed as to what he takes into his brain-attic. He will have nothing but the tools which may help him in doing his work, but of these he has a large assortment, and all in the most perfect order. It is a mistake to think that that little room has elastic walls and can distend to any extent. Depend upon it there comes a time when for every addition of knowledge you forget something that you knew before. It is of the highest importance, therefore, not to have useless facts elbowing out the useful ones."

Watson : "But the Solar System!" I protested.

Sherlock Holmes : "What the deuce is it to me?" he interrupted impatiently: "you say that we go round the sun. If we went round the moon it would not make a pennyworth of difference to me or to my work."

Looking back, is Sherlock Holmes not right? Should we also limit ourselves to what is useful for our line of work? Or the skeptic in you says the brain has every right to store any kind of information. I reflect on the above paragraph and feel Mr.Holmes was probably right. What do you think?