(For new reader and those who request 好友请求, please readmy 公告栏 first) 

Recently,I atteed the following seminar at Harvard:

“The Aha! Moment: From Data to Insight”

Dafna Shahaf of Stanford University

Abstract provided by the speaker

The amount of data in the world is increasing at incredible rates. Large-scale data has potential to  transform almost every aspect of ourworld, from science to business; for this potential to be realized, we must turn data into insight.  In this talk, I will describe two  of my efforts to address this problem  computationally:

The first project, Metro Maps of Information, aims to help people understandthe underlying structure of complex  topics, such as news stories or research areas. Metro Maps are structured summaries that can help us understand the information landscape, connect the dots between pieces of information, and uncover the big picture.

The second project proposes a framework for automatic discovery of insightful connections in data. In particular, we focus on identifying gaps in medical knowledge: our system recommends  directions of  research that are both novel andpromising.

I will formulate both problems mathematically and provide efficient, scalable methods for solving them.  User studies on real-world datasets demonstrate thatour methods help users acquire insight  efficiently across multiple domains.

Here is what I got out of this interesting talk:

Let me givean example from personal experience. I work in the area of “systems, control,and optimization”. Before I retired, how do I keep myself current and on the cutting edge of knowledge in the area. Well, these are my sources:

1. Going to conferences to listen to talks and conversing with my  colleagues face-to-face

2. Scan and/or read journals I subscribed to

3. Informal e-mail andother forms of communication with a subset of colleagues with similar interests

4. Unsolicitedcommunication from others wishing to attract my attention to their work

5. Supervising theses ofstudents and learn from their work

6. My own research work

7. Other ad hoc means (such as reviewing papers and books. I thank reader of comment [1] for making this explicit)

Theabove is a very time consuming process. I’d say 75% of my times are consumed   by these  endeavors from which I advance the knowledge of the subject  area. Besides pure teaching, you can say that is my job! Now because of “big data”,  Others such as the speaker claim it is  actually possible to automate much of tasks  #1-#7. The  question is “how much of the human judgment required for  distilling “INSIGHT” from  the data present in these tasks can be replaced by a machine  algorithm? ” The speaker report some successes with her algorithm in many  experiments in all areas of science, business, daily news reporting and sociology. On the other hand, I am also reminded of the quotes of one of my classmates, Stuart Dreyfus– professor emeritus of UC Berkeley

"Expertise ispattern discrimination and association based on experience. It is intuitive.There is no evidence you can reduce it to rules and theory. Hence, Artificial Intelligence probably  can't be produced using rules and principles. That's notwhat intelligence is."

But then again, neural network scientists would say that is why they think  human brain is a giant neural network which learns, adapts, and instantaneously decides by what we call intuition. Given time and money, they will triumph as the ultimate winner in AI (Note: they have been saying that for much over half a century since McCulloch and Pitts in 1943).

It is not likely the issue will be resolved in my lifetime.

Note added 4/5/2014 I thank Professor Weibo Gong of U Mass Amherst for providing this additional reference on the topic

http://www.newyorker.com/online/blogs/elements/2014/01/the-new-york-times-artificial-intelligence-hype-machine.html