Tuesday, July 08, 2014


Thursday, February 28, 2013

Mr. Jalil's petition

Original Bengali version by Humayun Ahmed
More about 1971 war atrocities

--------------------------------------------------------------------------------------------------------------

He said, with a smile, "I am the father of two martyrs. In 1971, both of my sons were killed."
I looked at him in surprise.
The guy was ordinary looking. Almost 60 years old.
He seemed strong enough compared to his age.
He was sitting straight, with a sharp gaze in his eyes.
He didn't have any glasses, which means his sight was good.
I asked, "How can I help you?"

He kept sitting, like before.
"I found a dead body," he said in a simple tone, "..buried it in Malibagh.
My youngest daughter lives in Malibagh."

Saturday, February 16, 2013

Theory induced blindness

An old Daniel Kahneman article on theory induced blindness 

"Once you have accepted a theory, it is extraordinarily difficult to notice its flaws. "

And we are never going to notice flaws if every time the theory fails to explain something, we keep saying "there's something mysterious about it that is beyond my intelligence" instead of questioning it.

Thursday, June 07, 2012

Choice

My religious friends, who are very concerned about what a woman should wear as if that is the biggest problem that the world needs to solve, share posts where men are saying how making women stay at home would restore peace in the world.

These guys, they talk as if a woman belongs to another species. Things do not affect a woman the same way they affect them. Women do not feel warm wearing hijab in 100 degrees. Women are not supposed to love to do something else other than raising a family or being dependent on a man. And women are not intelligent enough to decide what to do with their life or what to wear. Taking care of family or raising a kid or thinking about balancing work and home are female problem, not human problem.

They say women are given "choices." "Good" women "choose" to cover herself and "choose" to stay home and "choose" not to be ambitious. Choice is a really interesting word.

Monday, May 28, 2012

Want to change writing style? Do not use Google translator

Every time we give our authorship recognition talk, someone will say, "just run your text through a machine translator, it'd change your style!"
Same comments were made at IEEE S&P after the authorship recognition talk ("On feasibility of internet scale authorship recognition").

One of my lab mates experimented with translation effect on authorship recognition and found that translation cannot be used for writing style change.
The reason is that a good translator would keep the style intact and a bad translator would change the style so much that it would distort the original meaning completely.
A good writing style anonymizer should change the style while keeping the meaning of the text intact.

The paper on translation effect is still under submission.
Mike mentioned this a little in his talk at ccc, 2009.

Authorship recognition of multi-authored document

There are very few works on this topic.
Most recent work is by Moshe Koppel and his team "Unsupervised Decomposition of a Document into Authorial Components." In this paper they analyzed Bible as a multi-authored document and decomposed chapters into two sets, chapters written by Jeremiah and chapters written by Ezekiel.  They used synonym usage to distinguish two authors.

The authors said this approach could be used iteratively for more than 2 authors, but you need to know the number of authors.

There are couple of limitations of this work. They experimented with the Hebrew version of Bible, there were only two authors and they decomposed the text chapter-wise.


Monday, April 30, 2012

The Secret life of Pronouns

Ever noticed how you try to mimic the person you are interested in?
I know I do it, but I didn't know that it's true for everybody.

Pennebaker and his team did a study on speed dating conversation and found that our language changes when we are around people we like; in the speed dating case, "when the language style of two people matched, when they used pronouns, prepositions, articles and so forth in similar ways at similar rates, they were much more likely to end up on a date."

 Full story

Tuesday, April 24, 2012

How to fool the ETS essay rater?

Construct your prose with superfluous dunandunate words or phrases, although incoherent and ambiguous, which might cause xenolexica in human grader but impress the e-rater nonetheless. This is how Les Perelman fooled the e-rater.

Friday, April 06, 2012

My "cool" craigslist scammer

Sarah Lawrence (sarah.law03@yahoo.com) wrote:
Hi,

Thank you very much for the response but before we move forward,i will like you to know the full information about me. I was born and raised in Wyoming,i am quiet and easy going person to live with. I'm 26 years of age... I am extremely clean,quiet and respectful, I am sensitive, intelligent and am easy to get on with. ..I am a faithful and committed person, who is very considerate of other people's feelings. I don't drug and i dont smoke or drink.. I enjoy going out with my friends to movies, plays, parks, hikes, anything to do with the ocean and horses but i'm cool living with people.

Right now, am working for a Non Government Organization on a program on children with orphans and heart related probs.My next program/seminars will be in your town and I will need a room to stay for these seminars, so I want to secure a room before my arrival.

I must confess that i'm comfortable with the area of the house/apt and i am ready to pay for the room. I will like to make the payment for the first month ahead of my arrival, and then i will make subsequent payment thereafter as i plan to stay for 6-12months.I have told my boss about this and he promised to make sure that he will meet up with the payment so that I can depart from here as scheduled.He also said I will have the advantage to make payment for deposit .I should arrive to your place on or before the 24th of April(Willing to pay for the whole month of April),so pls keep the room for me till my arrival.He also said my mode of payment will be the Most acceptable,safer and reliable means of payment which is United state certified money order/Cashier Check so get back to me with your name and address so that i can send you the payment as early as possible.

Kindest Regards,
Sarah Lawrence.
===========================================================
I wrote:
Hi Sarah,
Sorry I cannot rent apartment from April. The earliest you can move in is --.
===========================================================

Sarah Lawrence (sarah.law03@yahoo.com) wrote:
Hi,

Thanks for the response...I am cool with everything and i will like to pay you the first month rent before my arrival...You can therefore get back to me with your name and address so that my Boss can send you the payment as soon as possible Ok..

Thanks again
Sarah Lawrence.

===========================================================

So cool!
So I wrote:
You know, scamming (or whatever you are trying to do here) needs some more intelligence than this.

===========================================================
Or may be not!
I wonder how many people fall for this and what exactly do they do after they get the address.
I guess, they were going to ask for my account number for direct deposit :P.
I should have gone few more steps to see what happens!
May be next time :D.

Other incidents of the same scam are here and here.

Wednesday, March 14, 2012

"The idea of capturing the intelligence of the readership -- that's a joke."

Nick Denton, the founder of Gawker media, made that remark on cnn.
Of course it is, just as raw gold in gold mine is dust!

I am currently analyzing comments from jezebel.com, the most commented site among the Gawker media sites, to see if we can predict comment quality using machine learning. It is actually surprisingly simple to translate most of the commenting guidelines into features that a machine learning algorithm would use to assess comment quality. Though some features are harder to determine than the others, for example, finding relevance of a comment with the original post or funny comments.

I made this graph with 2000 comments from jezebel that shows the timing effect on crowd rating. The x-axis shows time difference of the comment with the original post and y-axis shows number of comments made within that time interval. That is, 100 comments within 20-30 means that 100 comments were made within 20-30 minutes after the post was published.

On jezebel best quality comments are "promoted" by the trusted users. This graph shows that most  highly rated comments were made within 2 hours of the posts. So if you are late to make your awesome comment, there will be hardly any crowd to judge it.

I found two things that online communities value the most in comment assessment are timing and reputation. If you are late at making your comment, no matter how high quality the comment is there won't be any crowd to read your comment to rate it. If you have a good reputation of making good comments or if you have lots of friends in the community, your comment that says "wow!!!" would be "liked" by 100 people. The comment rating process seems democratic but it's very biased.

Automated comment filtering research has been done on many communities, yet I don't think any of the online communities uses it. Does anybody know of any community that uses some kind of machine based moderation?

Wednesday, February 29, 2012

Rethinking my Facebook friend acceptance policy

My Facebook-friend-acceptance policy is this.
Actually more restricted than that. If I don't share my personal detail with you in real life, I don't want to share it with you on Facebook.

The Facebook data team recently published this study which basically tells that if you want to receive interesting diverse news from your social network you should add people you hardly talk to. People with whom we have strong ties are similar minded, so they share things that we'd have shared. If you social network only contains similar minded people it'd become an "echo chamber." The more diverse your network is the more diverse and novel information you'd get from it.

I find the study very interesting. People with weak ties are good for getting interesting information, but if I have too many of them then I'd have to add extra privacy settings on my photo albums and status updates or be extra cautious about things I share. I'm not sure if people use facebook for getting news, it's an added benefit. News or no news, with my friends literally staying all over the world, I'd use Facebook anyway.

Thursday, December 08, 2011

I want to know if you can see Beauty
even when it is not pretty
every day.
And if you can source your own life
from its presence.


Sunday, August 07, 2011

I watch the stars from my window sill
The whole world is moving, and I'm standing still

Sunday, July 24, 2011

How far I am from home!


How far I am from home, originally uploaded by sheetal_shundori.

Feeling homesick