Tips for writing the thesis

And here's another link. This one has some useful tips when you (start) writing your thesis, including typesetting, acronyms, and more. See http://staff.science.uva.nl/~vdham/projects/typesetting-thesis.html.

On the history of evaluation in IR

I'm currently writing my PhD thesis and this paper is one of the gems I found: "On the history of evaluation in IR", S. Robertson, Journal of Information Science 34(4), 2008. You can find it here: http://jis.sagepub.com/cgi/content/long/34/4/439 (there's also a citeseer version for those without access I believe).

I think it's a must-read for any PhD student doing IR since it covers the fundamentals (and history!) of evaluation in IR. It's also written in a light and sometimes even funny style.

Enjoy :)

Don't be a Spock - Adjust your monitor height

Some practical guidance for all you guys working on your PhDs: http://www.xadamdx.com/2010/08/dont-be-spock-adjust-your-monitor.html.

3 shell scripts to improve your writing, or "My Ph.D. advisor rewrote himself in bash."

Fun read!

The hardest part of advising Ph.D. students is teaching them how to write.

Fortunately, I've seen patterns emerge over the past couple years.

So, I've decided to replace myself with a shell script.

See: http://matt.might.net/articles/shell-scripts-for-passive-voice-weasel-words-duplicates/.

Your research career: Academia or Industry?

Dr. Chistoph Bartneck, an assistant professor at TU Eindhoven gave this career talk (~12 min.) at the HRI conference in Osaka recently where he discussed the merits and cons of choosing between academia and industry after grad school. It's a truly 'sobering' talk that's extremely relevant to all of us so I had to pass it on. While many of us are generally aware of these differences, the power of the talk comes from tying it all together neatly.

You can watch the talk here.

Knowing your statistics is a fact of scientific life...

...and it's no exception for us PhD students. Below are three resources that have made statistical concepts and their application(s) truly accessible and intelligible to me (though I admit statistics was never really my strength).

Introduction to Statistics and Multivariate Statistics by David W. Stockburger from Missouri State University and Statnotes: Multivariate Statistics by G. David Garson from North Carolina State University.

Some good advice from the guys at Cornell

There's a long list of links with tons of good advice for PhD students at this site: http://www.cs.cornell.edu/People/czars/mentor/. I loved the one about getting your abstract rejected by Mary-Claire van Leunen and Richard Lipton.

Some light reading

If you get to that stage where you need to start The Writeup (aka your PhD dissertation), you might be interested in some tips. See How to Write a Ph.D. Dissertation.

:-)

You and Your Research by R. Hamming

Richard Hamming, a mathematician known for the Hamming Code, the Hamming distance along with numerical methods gave an interesting, and inspiring speech at Bell Labs 7 July 1986: You and Your Research. For those who don't have time do go through the entire speech, a condensed version can be found in: A Stroke of Genius: Striving for Greatness in All You Do.

It stroke me when I read that researchers do not cope with problems for which they don't have an "obvious" approach:

There are many right problems, but very few people search carefully for them. Rather they simply drift along doing what comes to them, following the easiest path to tomorrow. Great scientists all spend a lot of time and effort in examining the important problems in their field. Many have a list of 10 to 20 problems that might be important if they had a decent attack. As a result, when they notice something new that they had not known but seems to be relevant, then they are prepared to turn to the corresponding problem, work on it, and get there first.

I also found comforting statements from the ``big guys'' that one doesn't have to be a genius for making substantial contributions:

Newton said, ``If others would think as hard as I did, then they would get similar results.''

Edison said that genius was 99% perspiration and 1% inspiration.

and let's conclude with the Google Scholar tagline, from Newton:

"If I had seen farther than others it is because I stood on the shoulders of giants."

Happy reading, and researching!

Wolfram Alpha vs. Google

I just read the wired review of the Wolfram Alpha search engine on http://www.wired.com/epicenter/2009/05/wolframalpha-fails-the-cool-test/. They follow an opinion that I've read quite a lot - that Alpha is not going to be serious competition to Google.

Now that I've tried Alpha - I don't think competing with Google is the point really (independently of how Wolfram is trying to position the system himself, I guess there's some PR involved as well).

I agree that Alpha is not very useful as a general purpose search engine. It does different things, for example specify and visualize relations between things. For example, it allows you to compare things. "Amsterdam vs. Berlin" tells you that the number of people living in the metropolitan areas of both cities is quite similar (3 vs. 4 million), and that the flight time between the two cities is about 40 minutes. As people start trying Alpha more, I suspect that they will find a lot of uses, even ones unintended by the creators at this point.

Template for your PhD Thesis

Ready to start writing your PhD thesis? Check out the LaTeX template maintained by our original 'parent' institution: the Insititute for Logic, Language, and Communication. You can find it at http://www.illc.uva.nl/SupportandInfo/illcdiss.html

RSS & Co.

One of my students asked me how I keep up with conferences and journals.
Here are the resources I use:

RSS feeds
I subscribe to the blogs of a few people in my research area, and to ILPS PhD Resources of course. But I find RSS feeds the most useful for keeping up with journals. Some of the journals I subscribe to are:

• Computational Linguistics, from MIT Press (current issue | RSS feed)
• Information Processing & Management, from Elsevier (current issue | RSS feed). This is the only journal RSS feed that doesn't include article abstracts, which makes it a lot less useful than the other ones.
  
• Information Research, privately published (current issue | RSS feed)
  
• Information Retrieval, from Springer (current issue | RSS feed). I subscribe to a lot of journals from Springer, but I'm too lazy to look up all the links. Here are just the names: "Cognition, Technology & Work", "Computer Supported Cooperative Work", "Data Mining and Knowledge Discovery", "Minds and Machines", "Pattern Analysis and Applications", "User Modeling and User-Adapted Interaction", "World Wide Web".
  
• JASIS&T (Journal of the American Society for Information Science and Technology),
  from Wiley InterScience (current issue | RSS feed)
• Library Quarterly, from Chicago Journals (current issue | RSS feed)
  

mailing lists
The usefulness of these varies, some mainly have conference announcements, others have lively discussions. I'm not posting the addresses, most of these you can find on the web, or email me.

• Corpora-List
• SIG-IRList
• SIKS (I think people get subscribed to this one when they join the SIKS school)
  
• um (User Modeling, through Yahoo! Groups)
• webir (Yahoo! Groups)
  

Google Calendar
There's a Google calendar called "Information Retrieval & Web Mining Conference Dates". It includes submission deadlines and conference dates. I'm not sure how to create a link to this, but you'll find it if you search for the name in Google's public calendars.

What do you use to keep up with these things? What mailing lists and RSS feeds do you subscribe to?

Science 2.0

Scientific American published an article about Science 2.0 and how to make use of new technologies available. Is publishing raw results to the general public dangerous? Will the referencing scheme work? Will the ideas be copied? Or will there be a trajectory similar to Open Source?

Machine Learning followed up the article with a post.

Information Retrieval (study) resources

If you're looking for a way of getting to know the background of IR, there's a plethora of options (online) available. From the classic book by Keith van Rijsbergen (1979), the upcoming IR book by Manning and Schuetze (who previously wrote the classic "Foundations of Statistical Natural Language Processing") to various lecture pages of IR courses taught around the world. For example, David Carmel's course page at Haifa, James Allen's course page at UMass, Jamie Callan's course page at CMU, or ofcourse the course taught by Maarten de Rijke (which is more general and centered around "internet information").

Robin's advice

Here's a nice post from Robin Aly (doing his PhD at the Univ. of Twente): http://robin-twente.blogspot.com/2008/02/lessons-learned.html

Lectures on "Search Engines: Technology, Society, and Business"

This lecture series on search ran at UC Berkeley in fall 2005. Although some of the material may already be out of date and generally doesn't go very deep, it's still a good (broad) overview of topics related to search and provides insights from different perspectives. They got some people from industry and the videos of the lectures are good quality.

Here's the course schedule with links to individual lectures: http://www2.sims.berkeley.edu/courses/is141/f05/schedule.html

Style: Ten Lessons in Clarity & Grace

My thesis adviser insisted that I read this book and it turned out extremely helpful. Joseph Williams' "Style: Ten Lessons in Clarity and Grace" demystifies the process of writing clearly and provides exercises and ideas that help improve your writing no matter how well (or not) you write right now.

Focus of the book is the question of what makes a sentence easy or hard to read. Only a small part of the answer relates to grammatical correctness. The larger part of the book discusses ideas such as cohesion, coherence, and emphasis, how these can be assessed, and how sentences can be rewritten to communicate ideas clearly.

There are many editions of this book. At some point it was split up into three different books: "Style: (Ten) Lessons in Clarity and Grace", "Style: The Basics of Clarity and Grace", and "Style: Toward Clarity and Grace". I have the fifth edition of the first one, which can be bought used on amazon.com or half.com. I'm not sure how the other two relate to this one.

IR datasets on the web

On behalf of Manos:

Dear ilps-ers,

I found a nice post listing a lot of Large DataSets available on the web:

http://www.datawrangling.com/some-datasets-available-on-the-web.html

There is also a joint initiative to build a list of large datasets which can be found at: http://theinfo.org

You may already have heard most of the datasets but I thought this may prove handy as they are listed altogether.

Manos

How to Write a Good Paper: Some Suggestions

This month Krysztof Apt from the Dutch Center for Mathematics and Computer Science (CWI) gave an interesting presentation on how to write a good article. The presentation is now available on our video seminar search engine, ISLA-TV. The direct link to the video is here.

The abstract of the talk:

In this talk I shall go in detail into the writing process of a scientific article. In particular, I shall provide some suggestions how to organize the article, in what order to write its parts, how to organize and break down the proofs, how to link the results together, and what pitfalls to avoid.

CiteULike and more

CiteULike is in my opinion probably one of the finest web based tools for doing research. We even have our own group page there, in which you'll find a very broad scope of interests. Another one of my favourites is Diigo, which is a best-of-breed combination of Google Notebook and del.icio.us. It does much the same things as del.icio.us, but with this tool you can also highlight and annotate parts of documents found on the web and keep notes regarding those snippets. These will be stored, such that when you visit the page again, the highlighted sections will immediately stand out. Using their browser plugin/toolbar, you can even simultaneously add links and tags to other tagging software (Simpy, Furl, del.icio.us, Yahoo!, and Connotea to name just a few) and automatically create links in your own local bookmarks folder.