So I was pretty surprised to see many of my colleagues receive flowers and other gifts from their graduate students on Friday. Taesik was treated to lunch by his students and brought home a box of chocolate (and I ate ~90% of it). And as I saw some Twitter messages saying “I got these flowers from my students”, I was pretty jealous, I have to say. At the same time, I wondered what it all means. Do Korean students feel more appreciative of their graduate advisors’? Do the advisors take on more responsibilities, other than purely academic advice, for their students?

There are definitely cultural differences between the American and Korean advisor-student relationships. Certainly the language shows clear differences. American graduate students (at least in computer science) address their advisors by first name. In Korea, my students call me “Professor”. Students use an honorific form of the language to their advisors, and the advisors do not. I feel that the formality of the language acts as a barrier against really open discussions and exchanges of idea between the student and the advisor. In general, the Korean advisor-student relationship seems much more vertical, and hence, I think, the ballyhoo on teacher’s day.

Anyway, as I was pondering these thoughts, I felt relieved that my students did not shower me with flowers and presents. I felt that I am not really a “teacher” to my graduate students, as they seem to teach me more than I teach them. I also felt that I don’t really deserve the label “teacher” yet, since I have not graduated any students, and in perhaps ten years, by the time I have produced a few Ph.D.s, I will feel more deserving of the label “teacher”.

But as I walked into my office this morning, I found flowers and balloons on my desk, and messages of appreciation and love on my board. And all my “deep” thoughts about what it means to be a teacher just disappeared into the cloud. I am just enjoying the flowers, and most of all, the love of my students. As I read somewhere in Twitter this morning, people crave power because ultimately, power brings them love and attention. I hope that my students love me not only because I have the powers to sign or deny their theses, but because they truly appreciate what I try to do.

I love my students.

(I am sure I left out more than a few here. Please remind me!)

Old Friends

Krzysztof Gajos (Harvard)
Christine Alvarado (Harvey Mudd)
Jaime Teevan (Microsoft Research)
Rob Miller (MIT CSAIL)
Max Van Kleek (MIT CSAIL)
Aaron Adler (MIT CSAIL)
Maurice Chu (PARC)
Frank Bentley (Motorola)
Tracy Hammond (Texas A&M)
Stuart Schechter (Microsoft Research)
Jaeyeon Jung (Intel Research)
Taemie Kim (MIT Media Lab)
Erik Stolterman (Indiana University)

New Faces

Jimmy Lin (Google)
Greg Little (MIT CSAIL)
Elena Agapie (Harvard)
Hiroshi Ishii (MIT Media Lab)
Ed Chi (PARC)
Gene Golovchinsky (FX Pal)
Miles Efron (UIUC)
Julia Grace (IBM Almaden)
danah boyd (MSR New England)
Shamsi Iqbal (Microsoft Research)
Joan DiMicco (IBM Cambridge)
Michael Bernstein (MIT CSAIL)

It was great meeting you all, and I look forward to seeing you at CHI 2011 (or sooner).

On Sunday, my student Dongwoo Kim will present our paper at the Microblogging workshop. Another student, Sunjun Kim, has a work-in-progress poster. And I am a co-author of a paper for the “Know Thyself” Personal Informatics workshop. Finally, I will be chairing a session “Cooking, Classrooms, and Craft” on Thursday morning.

Of course, I am looking forward to attending sessions, meeting friends, and generally having fun.

Here are our papers:

Dongwoo Kim, Yohan Jo, Il-Chul Moon, and Alice Oh. Analysis of Twitter Lists as a Potential Source for Discovering Latent Characteristics of Users. Workshop on Microblogging at the ACM Conference on Human Factors in Computer Systems (CHI 2010).

Younkyung Lim, Alice Oh, Tekjin Nam, and Kee-Eung Kim. Personal Informatics for Discovering Human-Centered Lifecare System Opportunities. Workshop on Know Thyself at the ACM Conference on Human Factors in Computer Systems (CHI 2010).

Since we have been working with topic models for the last six months, I thought David Blei at Princeton would be the best person to visit. He accepted my request, and on top of that, invited us to give a talk at his machine learning reading group. Although we don’t yet have fancy results, we gave a presentation that summarizes our research goals, preliminary results, discussion questions, and future plans. We also had a separate meeting with Dave which led to helpful discussions and neat ideas. Dave was a gracious host and answered lots of our questions even though he had a very jam-packed day because it was the visit day for Princeton CS new admits. It was a great experience for us, and we will follow up on our discussions with Dave by trying out some new ideas for LDA evaluation and topic-sentiment model.

Here are the slides from the presentation. (Yohan and I made & delivered the talk, Dongwoo and Hyunjong contributed to the results in the talk.)
Talk @ Princeton, March, 2010

The readings associated with the presentation are:

J. Chang, J. Boyd-Graber, S. Gerrish, C. Wang, and D. Blei.    Reading tea leaves: How humans interpret topic models. Neural Information Processing Systems, 2009.

C. Lin and Y. He. Joint Sentiment Topic Model for Sentiment Analysis. CIKM 2009.

If you’re interested in learning about LDA, Dave’s tutorial is great.

Call For Papers

We solicit research papers, works-in-progress, and position papers. Click here for the CFP in PDF.

Important Dates (US EDT)

Submission Deadline: May 1, 2010
Notification of Acceptance: June 8, 2010
Final Paper Due: June 15, 2010

1. Workshop Overview

This is a workshop at IEEE International Conference on Social Computing (SocialCom-2010), and the goal of the workshop is bring together researchers from various disciplines to discuss ways to find synergies between text analysis and network analysis. The workshop will contribute to Social Computing by

1. initiating a dialogue about and an in-depth look at the methodologies used in social network analysis and text analysis
2. discussing holistic approaches for combining network and text analysis
3. sharing and defining needs for data sets that include content and metadata
4. illustrating informative data analysis results from the combination of text and network analysis
5. presenting novel applications combining text and network analysis including their validations

Click here for more information.

I did this in part because Taesik is suffering from a wisdom tooth extraction procedure, and it was a win-win-win situation. I gave him a couple of days to recover from the operation without having to give a lecture. I had a good time preparing for and delivering the lecture, and at the same time, I got to organize my thoughts about applying the ideas and methods used in our lab to health information. And I think, for the most part, students appreciated this change of perspective, from a systems engineering point of view to a computer science point of view.

I spent quite a long time preparing the presentation, so I want to share it with everyone here. Note that it is not a very technical talk. It’s intended to serve a general audience.

The Social Web and Health Information: A Computational Perspective

P. D. Turney and P. Pantel (2010) “From Frequency to Meaning: Vector Space Models of Semantics”, Journal of Artificial Intelligence Research, Volume 37, pages 141-188

http://www.jair.org/papers/paper2934.html

During the Olympics, Yuna’s coach and Mao’s coach, their faces showed different emotions. I think that it was not just because of the differences in Yuna and Mao’s performances at the Olympics. I think their faces reflected the very different styles of coaching.

As for me, my biggest concern is finding my coaching style. Research is hard, but advising students is much much harder.

It is also much more rewarding.

In fact, when I was a PhD student, I did not think much about a career in academia. I somehow stumbled upon this profession, but now I am very glad that I am a professor in a research-oriented university with many opportunities to advise graduate students.

For the most part, the results of our research do not make big impacts in the real world, but advising students can and do make big differences.

I try to talk to other faculty to get advice on advising students. But there seems to be no right answer. Every student is different, every advisor is different, and every interaction is different. There seems to be no way to get a training set and expect it will be from the same distribution as the testing set. Nevertheless, learning must occur. How???

Fortunately, we have intelligence that does not rely solely on statistics. You just take on each interaction with each  student, try the best you can, and hope to get some positive results.

And with experience, I hope, I will get better with those interactions. I am not sure, though, whether I will ever feel qualified to be their role model. For that, I will have to try harder. Anyway, for now, every day is a struggle just to make sure I do not make unrecoverable mistakes.

Of course, I enjoy it immensely!

Spring semester is starting!

There has been much change since my last post, over a year ago…

The biggest change in my life is the birth of my second child.

He’s certainly a bundle of joy!

Unlike our daughter, who is very cautious, thinks deeply, and takes a while warming up to new faces, this baby seems very happy all the time, even with new faces.

It’s a bit challenging to manage time with a newborn, second grader, and a more-than-full-time job as a tenure-track junior faculty, but I am very very lucky to have my parents helping out with our two kids. Very grateful indeed!

Good luck to all academics, faculty and students, starting a new semester!

Artificial intelligence: An instance of Aibo ingenuity
by Michael Littman