Semantic Web takes off ….

April 24th, 2007

or does it? Decide for yourself …

Yesterday I got one of those glossy multipage conference brochures in the mail. I was intrigued, because instead of being for Oracle or SAP or Office or some such, it was for an event dubbed “Semantic Technology Conference” that described itself as “The foremost place to learn about the commercialization of Semantic Technologies”.  

The brochure described many interesting talks, mostly on uses of the semantic web technologies. Looking at that, it would seem that the semantic web is finally taking off! There were even some talks from researchers I recognized on technology issues (Deb Mcguinnes, Amit Sheth, Stefan Decker, Mike Dean, John Sowa, Vasant Honavar and others). A UMBC (and Ebiquity/LAIT) alumna Melli Annamalai was also represented in the list of speakers.

On the other hand, my colleage Tim Finin has a contrarian take — perhaps conferences like these represent the peak of the hype curve for the semantic web, after which comes the inevitable fall.

Regardless of this, one interesting new “buzzword” I got from the brochure is Web 4.0, where the semantic web meets ubiquitous computing. The accompanying chart sets it start date to be 2015. Perhaps we should go out and get patents on all our work done in this space in the last 5 years, and hope to get rich a decade from now 🙂

Making machines, and computer games, intelligent

April 23rd, 2007

People are remarkably good at “muddling through”, at not getting completely stuck, at continuing to make some progress even when things don’t go as planned. Machines, on the other hand, tend to fail in dramatic fashion when they are faced with unusual circumstances.

UMBC professor Tim Oates is working on building more robust intelligent systems through metacognition, which is the ability to think about your own thinking. Meta-cognitive systems can notice when things are not going well (as opposed to just plowing ahead with whatever they were doing, which is what most systems do today), reason about possible sources of the problem, and try various repairs.

Professor Oates and his colleagues have constructed a meta-cognitive computer player for the tank game Bolo that learns from its mistakes and adapts its knowledge, tactics, and strategies as it faces new challenges in the game and more capable human or automated opponents. For more information see the ALMECOM pages, some of the detailed papers on metacognition in Bolo and the UMBC games, animation and interactive media blog.

Metagognitive bolo tanks

On Wikipedia’s article on the Virginia Tech Massacre

April 23rd, 2007

Today’s New York Times has an article, The Latest on Virginia Tech, From Wikipedia, on Wikipedia’s article about the Virginia Tech massacre that happened just one week ago.

Imagine a newspaper with more than 2,000 writers, researchers and copy editors, yet no supervisors or managers to speak of. No deadlines; no meetings to plan coverage; no decisions handed down through a chain of command; no getting up on a desk to lead a toast after a job well done. … From the contributions of 2,074 editors, at last count, the site created a polished, detailed article on the massacre, with more than 140 separate footnotes, as well as sidebars that profiled the shooter, Seung-Hui Cho, and gave a timeline of the attacks. According to the foundation that runs the various Wikipedias around the world, there were more than 750,000 visits to the main article on the shootings in its first two days, an average of four visits a second. Even The Roanoke Times, which is published near Blacksburg, Va., where the university is located, noted on Thursday that Wikipedia “has emerged as the clearinghouse for detailed information on the event.”

UMBC Digital Entertainment Conference 28 April 2007

April 23rd, 2007

The UMBC Game Developer’s Club will host the second Digital Entertainment Conference on careers in the computer game and interactive entertainment industry. The event will take place on Saturday April 28th from 10:30-5:30 in Lecture Hall 5 of the UMBC Engineering and Computer Science Building. It is free and open to the public.

The 2007 DEC will feature presentations from seven speakers from four Maryland game companies and cover a range of topics, including art, programming and production.

See the GDC site for a detailed schedule and biographic sketches of the presenters.

NYT on Twitter: From Many Tweets, One Loud Voice on the Internet

April 21st, 2007

Sunday’s New York Times has an article on Twitter, From Many Tweets, One Loud Voice on the Internet. The article covers the basics, e.g.

“Twitterers” send and receive short messages, called “tweets,” on Twitter’s Web site, with instant messaging software, or with mobile phones. Unlike most text messages, tweets — usually in answer to Twitter’s prompt, “What are you doing?” — are routed among networks of friends. Strangers, called “followers,” can also choose to receive the tweets of people they find interesting.

and mention some of the ancillary services, like Twittervision and Twritterholic, but not Geotwitter or Twitterment. :-(

The article was written by Jason Pontin, EIC of Technology Review, who tried Twitter and was somewhat conflicted:

My own experiences with Twitter were mixed. I quickly realized that decrying the banality of tweets missed their point. The only people in the world who might be interested in my twittering — my family, my close friends — were precisely the ones who would be entertained and comforted by their triviality.
    But I also strongly disliked the radical self-revelation of Twitter. I wasn’t sure that it was good for my intimate circle to know so much about my daily rounds, or healthy for me to tell them. A little secretiveness is, perhaps, a necessary lubricant in our social relations. I wondered whether twittering could ever have broad appeal.

I’m not sure I understand the conflict though — he’s right, I think, that Twitter can fulfill an ambient intimacy function, whether its in a personal (e.g., family) or professional (e.g., co-workers) domain. Most people choose twitter for one or the other and, once you have done that, you have total control over what you choose to reveal. The problem arises if your followers are not from the same social group, i.e., a mixture of family and co-workers. While one can solve this with multiple Twitter accounts, it might be interesting to explore a simple context mechanism that could easily be added to Twitter.

Twitter Social Network Analysis

April 19th, 2007

In the recent series of posts, we have presented Twitter Goolgle Maps mashup, a Twitter search and buzz tracking tool called Twitterment and analysis of geolocation information from the twitter dataset. By providing a neat API, Twitter has enabled researchers to get a better understanding of Microblogging.

In this post, I have used the Large Graph Layout (LGL) tool to visualize the social network on Twitter. Following is a graph constructed using contacts from about 25K users. Notice that there is a link connecting two users if either one has the other as a friend and hence it is an undirected graph (of about 250K edges).

Compare this to the following graph that is constructed using only users who are mutually acquainted. i.e. A knows B and also B knows A.

I find that visualizing such large graphs is quite a challenge and to glean meaningful information from it is even more difficult. However there are a few insights one can gain from this:

  • Interestingly, there are a number of users who are trying to win a popularity contest of some sorts! The complete list of users ranked by the number of friends they have is shown here.
  • A number of bloggers and (perhaps fake?) celebrity profiles have a huge fan following in Twitter. Here is a list of users ranked by number of followers.
  • The two graphs shown above look very different on account of the fact that users with public profiles get a lot of followers whom they might not really know and would hence never add them as an acquaintance (well, in most cases atleast). But to really understand what the differences are one would need to look at the community structure and properties of the two graphs.

Finally, for completeness, here is a list of users ranked according to their PageRank scores. It is noticeably similar to the rankings generated by Twitterholic. This can be explained by the fact that local metrics (like number of followers) in a social network are a good first order approximation of rank. Dr. Finin made me aware of research by social network expert Valdis Krebs, who uses “reach” as a measure in human social networks. Here a person’s reach is the number of other people that are within N links in the network where N is usually 1, 2 or 3 for human networks. So, Twitterholic rank for example is the case with N equal to 1.
[Thanks Eytan and Matt for suggestions on Graph Visualization tools. Related: Matt, Bruno’s posts on network visualization of Belgian bloggers]

links for 2007-04-18

April 18th, 2007
  • Ronald Arkin of the Georgia Institute of Technology, in Atlanta, is developing a set of rules of engagement for battlefield robots to ensure that their use of lethal force follows the rules of ethics.

CS Takes Steps to Bring Women to the Fold

April 17th, 2007

Today’s NYT has a story, CS Takes Steps to Bring Women to the Fold, on the decline in the number of women studying computer science and what some universities are doing to reverse the trend.

Women received about 38 percent of the computer science bachelor’s degrees awarded in the United States in 1985, the peak year, but in 2003, the figure was only about 28 percent, according to the National Science Foundation.
    At universities that also offer graduate degrees in computer science, only 17 percent of the field’s bachelor’s degrees in the 2003-4 academic year went to women, according to the Taulbee Survey, conducted annually by an organization for computer science research.
    Since then, many in the field say, the situation has worsened. They say computing is the only realm of science or technology in which women are consistently giving ground. They also worry that the number of women is dropping in graduate programs and in industry.

It’s a trend that we atill don’t completely understand, which means that our efforts to reverse it may or not be on target.

Global Distribution of Twitter Users

April 15th, 2007

I find Twitter’s global appeal is truly amazing. Just a few minutes of watching Twittervision/GeoTwitter and searching for terms on Twitterment will prove its popularity. The map below shows the distribution of Twitter users across the world.

This map was generated by using the Tweets (Twitter posts) aggregated by Twitterment from Twitter public timeline API. It consisted of over 300K Tweets from 35K users. Not all users specify their location, but from the 19K who do – we were able to resolve the geolocation for about 11K users (using Google API). Shown on the right are the top 10 Cities for Twitter usage.
Some interesting points:

  • As Steve Rubel recently observed, Twitter seems to be quite popular on the East coast.
  • US, Europe and Japan are places where most tweets come from. Japan being the hotbed for new Cell phone/SMS technology, has been very quick at adopting Twitter.
  • By offering multilingual support and local SMS short codes (40404 numbers) Twitter might be able to do better in markets like India and China which already have a significant high SMS adoption.

Another interesting example of language analysis in twitter was posted by Bruno Peters. Variations of a term’s popularity can be applied to derive market intelligence for a given demographic. Sudden occurrences of concepts (like “Mexico earthquake“) are potential indicators of trends and geographically specific events.

Also, check out Matt Hurst’s analysis of popularity of the term “Twitter”. It is finally Twitter’s users who are making it popular by evangelizing and getting their friends across the world to join in!

PS: Thanks everyone for your comments and Tweets on Twitterment. More updates shortly.

Cumulative advantage = preferential attachment?

April 15th, 2007

The NYT has an article by Columbia’s Duncan Watts entitled Is Justin Timberlake a Product of Cumulative Advantage? on a fascinating experiment on how markets chose popular items.

Conventional marketing wisdom holds that predicting success in cultural markets is mostly a matter of anticipating the preferences of the millions of individual people who participate in them. … The common-sense view, however, makes a big assumption: that when people make decisions about what they like, they do so independently of one another. But people almost never make decisions independently — in part because the world abounds with so many choices that we have little hope of ever finding what we want on our own; in part because we are never really sure what we want anyway; and in part because what we often want is not so much to experience the “best” of everything as it is to experience the same things as other people and thereby also experience the benefits of sharing. … The reason is that when people tend to like what other people like, differences in popularity are subject to what is called “cumulative advantage,” or the “rich get richer” effect. This means that if one object happens to be slightly more popular than another at just the right point, it will tend to become more popular still. As a result, even tiny, random fluctuations can blow up, generating potentially enormous long-run differences among even indistinguishable competitors — a phenomenon that is similar in some ways to the famous “butterfly effect” from chaos theory.

This isn’t too surprising and is related, partly, to the well known preferentiall attachment concept from the Barabási-Albert (BA) model for scale-free networks. In a graph model, preferential attachment means that the more links a node is, the more likely it is to get new links.

Watts and his collaborators at Columbia created MusicLab as an environment for running experiments to explore the phenomenon as described in this paper in Science:

Experimental Study of Inequality and Unpredictability in an Artificial Cultural Market, Salganik, M.J. and Dodds, P.S. and Watts, D.J., Science, v311, n5762, p854, 2006.

Hit songs, books, and movies are many times more successful than average, suggesting that ‘‘the best’’ alternatives are qualitatively different from ‘‘the rest’’; yet experts routinely fail to predict which products will succeed. We investigated this paradox experimentally, by creating an artificial ‘‘music market’’ in which 14,341 participants downloaded previously unknown songs either with or without knowledge of previous participants’ choices. Increasing the strength of social influence increased both inequality and unpredictability of success. Success was also only partly determined by quality: The best songs rarely did poorly, and the worst rarely did well, but any other result was possible.”

They ran two groups, in one subjects were asked to rate and optionally download music without any knowledge of how popular others found it. The other group was split into eight distinct communities and were also asked to rate and possibly download. Subjects in each community could see how often the songs had been downloaded, a measure of their popularity. The findings were (1) that knowing a song’s popularity effected the rating and (2) the different communities edned up choosing different highly-popular songs.

In all the social-influence worlds, the most popular songs were much more popular (and the least popular songs were less popular) than in the independent condition. At the same time, however, the particular songs that became hits were different in different worlds, just as cumulative-advantage theory would predict. Introducing social influence into human decision making, in other words, didn’t just make the hits bigger; it also made them more unpredictable.

I wonder how well their results can be explained by preferential attachment.

UMBC programs on Games, Animation and Interactive Media

April 13th, 2007

The computer game industry has become big business. The Washington Business Journal ranks the Baltimore/Washington D.C. area third in the number of computer game companies nationally and GameCareeGuide reports that starting salaries are very good for both programmers and artists. Recent UMBC graduates have gone on to work in a number of the leading companies in the area, including Breakaway Games, Firaxis, and Mythic Entertainment.

To address student interest and also the demand for talented graduates by the local game industry, UMBC is developing a new game development track as an option for students pursuing a B.S. degree in computer science, and a concentration in animation and interactive media for students pursuing a B.A. degree in visual arts. We’ve set up a web site for information on UMBC’s programs in Games, Animation and Interactive Media and are also running an associate GAIM blog.

links for 2007-04-13

April 13th, 2007