UMBC ebiquity research group Building intelligent systems in open, heterogeneous, dynamic, distributed environments
UMBC eBiquity Blog

Kaggle aims to host data-driven machine learning competitions

Tim Finin, 6:36am 3 February 2010

Kaggle is a site for data-related competitions in machine learning, statistics and econometrics. Companies, researchers, government and other organizations will be able to post their modeling problems and invite researchers to compete to produce the best solutions. The Kaggle demo site currently has three example competitions to illustrate how it will work and expects to host the first real one in March. Kaggle’s competition hosting service will be free, but the site says that it plans to “offer paid-for services in addition to its free competition hosting.”

 

UMBC global game jam live video feed

Tim Finin, 6:45pm 30 January 2010

Via Marc Olano: The Global Game Jam is into its second day at UMBC with 41 registered participants working on seven games. Keep up from home with our live video feed and games list.

 

Global Game Jam at UMBC, January 29-31

Tim Finin, 3:31pm 28 January 2010

UMBC will be the Baltimore site for the Global Game Jam. This is a 48 hour event, where teams from around the globe will work to each develop a complete game over one weekend. Last year, the UMBC site fielded five teams as one of 54 sites in 23 countries. This year promises to be even bigger, with 124 sites in 34 countries.

The Baltimore site and open to participants at all skill levels. It is not necessary to be a UMBC student to register. Thanks to generous support by Next Century , there is no registration fee for the Baltimore site, but you must register for this site in advance at www.globalgamejam.org. The jam will start at 5PM on Friday, January 29th in the UMBC GAIM lab, room 005a in the ITE building. At that time, the theme for this year’s games will be announced, and we’ll brainstorm game ideas and form into teams. Teams will have until 3pm on Sunday, January 31st to develop their games. We’ll have demos of each game and selection of local awards, wrapping up by 5pm Sunday.

Last year’s theme was “As long as we’re together there will always be problems”, and we had games developed using a combination of XNA, Flash, Maya, Photoshop, and the Unity Engine.

For more information, visit http://gaim.umbc.edu/jam/.

 

cfp: two special issues of the Journal of Web Semantics

Tim Finin, 2:40am 23 January 2010

The Journal of Web Semantics has announced two new special issues. Heiner Stuckenschmidt and Jeff Heflin are editing a special issue on web-scale semantic information processing with a deadline of 1 July 2010 for submissions. Grigoris Antoniou, Mathieu d’Aquin and Jeff Z. Pan are editing a special issue on semantic web dynamics with submissions due 31 May 2010.

 

How lotteries like Spain’s El Gordo enhance social ties

Tim Finin, 12:46am 3 January 2010

The economist has an article, Gamblers united, on Spain’s lotteries, like El Gordo (“the Fatty”) will will pay out €2.3 billion this year. What I found interesting is that this and other Spanish lotteries are events that enhance social ties.

“Loterías y Apuestas del Estado, the government agency that runs El Gordo and other lotteries during the year, encourages mass participation by dividing each €200 ticket into décimos, or tenths, which sell for €20. This, in turn, allows players to improve their odds by buying small shares in many tickets, often by forming syndicates with friends and colleagues. … All this has transformed the lottery from a glorified tax on the poor, as it is in most countries, into part of the social fabric. Sharing tickets at Christmas has become a way to reinforce social ties, says Roberto Garvía, a visiting professor at Georgetown University. The practice of forming syndicates, which initially started in the 19th century when lottery tickets became too expensive for working-class folk, has become a tradition among all classes. As one banker says, “I don’t want to be the only idiot who has to turn up to work if the office number wins.”

 

Spamassassin 2010 bug

Tim Finin, 3:18pm 1 January 2010

Shades of Y2K! Mike Cardwell reports on a rule in Spamassassin that judges any message sent in or after 2010 as “grossly in the future” and treats this as evidence of it being spam. I just checked and found that our mail server’s Spamassassin is using this buggy FH_DATE_PAST_20XX rule.

If you are using Spamassassin, or think your mail server might be, check the source of mail you have received today. Here’s an example from one of my messages this morning.

X-Spam-Checker-Version: SpamAssassin 3.2.5 ... on mail.cs.umbc.edu
X-Spam-Level: *
X-Spam-Status: No, score=1.6 required=5.0 tests=AWL,FH_DATE_PAST_20XX
  autolearn=disabled version=3.2.5
Received: from mail-yw0-f142.google.com (mail-yw0-f142.google.com
  [209.85.211.142]) by mail.cs.umbc.edu (8.14.3/8.14.3) with ESMTP
  id o01DjJUn011187; Fri, 1 Jan 2010 08:45:19 -0500 (EST)

If the message exceeds the local spam score threshold for, you may find a block with more details in your message header, like this example.

Content analysis details:   (6.1 points, 5.0 required)

 pts rule name              description
---- ---------------------- ----------------------------------
 3.4 FH_DATE_PAST_20XX     The date is grossly in the future.
-4.0 RCVD_IN_DNSWL_MED   RBL: Sender listed at http://www.dnswl.org/,
    medium trust [130.85.25.80 listed in list.dnswl.org]
 1.8 SUBJ_ALL_CAPS  Subject is all capitals
 0.7 MSOE_MID_WRONG_CASE  MSOE_MID_WRONG_CASE
 4.2 FORGED_MUA_OUTLOOK  Forged mail pretending to be from MS Outlook

As a workaround until your server updates Spamassassin, the points that the rule adds to a message’s spam score can be lowered to 0.0 in Spamassassin’s configuration file (local.cf) or your own user-prefs file.

score FH_DATE_PAST_20XX 0.0

 

UMBC wins 2009 Pan-Am College Chess Tournament

Tim Finin, 10:29am 31 December 2009

GM Leonid Kritz, UMBCUMBC won the 2009 Pan-American Intercollegiate Team Chess Championship yesterday with perfect score of 6.0 points. This year’s tournament was held in South Padre Island, Texas on December 27-30

This is from the UMBC press release:

“Capping a near-flawless performance over the past four days at the tournament in South Padre Island, Texas, the Retrievers topped a team from the University of Texas-Brownsville today to secure the title outright for the first time since 2005.

Today’s win completed a perfect 6-0 match record for the tournament, known as the “World Series of college chess.” UMBC’s dominant performance is all the more impressive given the high quality of the 28-team field this year, said Alan Sherman, director of the school’s chess program.

Today UMBC topped UT-Brownsville’s “B” team, 4-0, to complete the march to the title. But the Retrievers most of the hard work yesterday, winning decisive matches over two of the strongest teams in college chess. The Retrievers topped UT-Dallas, 3 to 1, in the early match, and then got past UT-Brownsville’s “into today’s action. The tournament also included teams from Yale, Princeton, NYU, Stanford and University of Chicago.

The Pan-Am is the most celebrated intercollegiate chess tournament in the Americas. Since its 1946 inception, dozens of universities throughout the Americas have participated. The tournament is open to any college or university team from North, South, or Central America.

Since 2003, the teams representing the top four schools in the Pan-Am have met again in the spring to compete for the President’s Cup in an event sponsored by the U.S. Chess Federation — the “Final Four of Chess”. In 2010 the University of Texas Brownsville will host the final four, UMBC, UTD, UTB and Texas Tech, in April.

You can get information on the tournament and the games at monroi.

 

Learning to love your robot

Tim Finin, 10:01am 22 December 2009

The new Scientist has an article, Learning to love to hate robots, on recent research on how humans and robots interact and ways to improve the relationships. The most popular robot in such “opposite relationships” is, of course, the little Roomba. Searching for roomba on Flickr produces more than 5000 pictures taken by their human friends.

“A six-month study of how Roomba affected households, conducted by Ja-Young Sung at the Georgia Institute of Technology in Atlanta, backs up that finding. “Some people saw it as a lifetime partner – they had a real emotional attachment to it.” Even those who returned to their previous cleaning routine didn’t blame the robot, instead saying it was their routine that was at fault.”

See their 2009 CHI paper, “Pimp My Roomba”: Designing for Personalization.

The little guy is pretty savvy — it knows how how to get ahead even if it doesn’t have the fastest cores on the block: manage expectations.

“One study by Jodi Forlizzi at Carnegie Mellon University in Pittsburgh, Pennsylvania, highlights how popular culture can affect a robot’s reception. People she introduced to Roomba, a robotic vacuum cleaner made by iRobot of Bedford, Massachusetts, compared it with their knowledge of robots that explore Mars, forming low expectations of Roomba’s abilities. But making a bad first impression seemed to help Roomba; it invariably surpassed expectations, helping people bond with their machine.”

See How Robotic Products Become Social Products: An Ethnographic Study of Cleaning in the Home.

 

Twitter hacked by Iranian Cyber Army

Tim Finin, 1:52am 18 December 2009

TechCrunch is reporting that Twitter is down due to an attack by someone claiming to be part of the ‘Iranian Cyber Army’. Since Twitter is now down, we can’t show a screen shot, but Techrunch reports that a similar defacement is live at mawjcamp.org.


Picture 1

    Iranian Cyber Army
    THIS SITE HAS BEEN HACKED BY IRANIAN CYBER ARMY
    iRANiAN.CYBER.ARMY@GMAIL.COM

    U.S.A. Think They Controlling And Managing Internet By
    Their Access, But THey Don’t, We Control And Manage
    Internet By Our Power, So Do Not Try To Stimulation
    Iranian Peoples To….

    NOW WHICH COUNTRY IN EMBARGO LIST? IRAN? USA?
    WE PUSH THEM IN EMBARGO LIST ;) 

    Take Care.

 

foaf:mbox_sha1sum considered harmful

Tim Finin, 11:52pm 17 December 2009

The foaf:mbox property is very useful since it is ‘inverse functional’ and can thus serve as an ID for a foaf individual. This lets us infer that two foaf profiles with the same mbox refer to the same person.

Since publishing your email address invites spam, many people use the foaf:mbox_sha1sum property instead of mbox. mbox_sha1sum is also inverse functional but doesn’t reveal your private information (i.e., email address).

Abell on developer.it has an interesting post, Gravatars: why publishing your email’s hash is not a good idea, that shows how to crack an MD5 hash of a person’s email address given a little information about the person. (note: The gravitar service supports globally recognized avatars.)

The idea exploits the fact that a few free email services (e.g., gmail, hotmail, yahoo, aol) account for a large fraction of email addresses and using a person’s full name, one can generate likely ‘username’ possibilities. Given an email hash and a persons first and last name, one can generate hashes of likely email addresses until a match is found.

Abell was able (!) to crack 10% of the email addresses for 80,871 stackoverflow.com users in an hour with a simple Haskell program.

The same attack can be used on foaf:mbox_sha1sum properties, especially since a foaf profile will very handily provide the other useful information about the person. Given the extra information available in many foaf profile (e.g., nick, school homepage) one might even expect better results.

As vulnerabilities go, this doesn’t seem like a very dangerous one. The use of mbox_sha1sum is usually justified as a way to avoid having your email address harvested by spambots. I doubt that spammers would think it productive to spend an hour of computing time to get 1000 email addresses.

 

Mathematical model predicts insurgent attacks

Tim Finin, 7:27pm 16 December 2009

Mathematical model predicts insurgent attacks

A paper just published in Nature, Common ecology quantifies human insurgency, describes a mathematical model that can be used to predict the the sizes and timing of violent events within different insurgent conflicts.

“We propose a unified model of human insurgency that reproduces these commonalities, and explains conflict-specific variations quantitatively in terms of underlying rules of engagement. Our model treats each insurgent population as an ecology of dynamically evolving, self-organized groups following common decision-making processes. Our model is consistent with several recent hypotheses about modern insurgency is robust to many generalizations, and establishes a quantitative connection between human insurgency, global terrorism and ecology. Its similarity to financial market models provides a surprising link between violent and non-violent forms of human behaviour.”

See also a note in Nature News, Modellers claim wars are predictable and this TED talk by one of the authors, Sean Gourley, on the mathematics of war.

The TED blog has more information and portions of an interview with Gourley.

gourley_schematic

 

Stanford AI pioneers reunite

Tim Finin, 8:39am 8 December 2009

Today’s NYT has a short essay by John Markoff, Optimism as Artificial Intelligence Pioneers Reunite, on a recent reunion of researchers from the early days of the Stanford AI Laboratory.

 







UMBC