NCSA Home
Contact Us | Intranet | Search

NCSA NEWS

News Home
Calendar
Images
Video on Demand
Subscribe to Our Newsletter
Frequently Asked Questions

 Intelligent Technology

IntelliBadge Used to Build Communities and a Smarter SC2002 Conference

When the National Center for Supercomputing Applications (NCSA) took on the IntelliBadge™ project for the SC2002 conference, no one knew quite what to expect. As it turned out, the demonstration of "smart" technology was a popular feature at the conference, held in Baltimore last Nov. 16-22. The IntelliBadge technology also shows much promise as a data-to-knowledge tool, useful to other conferences and to the business world.

The IntelliBadge demonstration used Savi Technologies' 600 Series RFID tags and readers to track people at the SC2002 conference by their interests, monitor the flow of people in and out of conference events, view conference statistics, and view visualizations representing the flow of people through the technical program and exhibit hall. A total of 890 SC2002 attendees participated in the project, about the maximum the project was set up to handle. IntelliBadge participants registered onsite at an IntelliBadge kiosk. They were then able to log into the system to check statistics and gather information, either at the kiosks or remotely through the IntelliBadge website.

NCSA's Experimental Technologies group developed the interactive database application that provided real-time information for the project and the visualizations used to display that information. The group is now analyzing the data obtained from IntelliBadge wearers. Among their preliminary findings are:

  • The maximum distance walked during the week-long conference was nearly 44 miles, while the average was about 6.9 miles. The lowest mileage logged was less than a half mile. The extreme lows were probably caused by participants who sometimes forgot to wear their badges and by those who spent a lot of time in areas not covered by the readers. The extreme highs might have been influenced by overlapping readers that inflated some mileage.


  • From the start of registration on Nov. 16 through the conference's end on Nov. 22, IntelliBadge participants used kiosk services 1,771 times onsite and 1,370 times remotely. The services were used most often to search for people (4,261 searches), check mileage statistics (2,623 searches), check current conference events (1,205 searches), and find area restaurants (721 searches). Of those users who looked for restaurant information, the "any food" category was searched most often, followed by "seafood" and "American."


  • The average user spent 3 minutes 38 seconds filling out their IntelliBadge registration and profile. The typical kiosk session lasted 1 minute 53 seconds and the total time spent by all users at the kiosks was 83 hours 7 minutes.


  • On average, attendees spent 7 hours 55 minutes on the exhibit floor and 3 hours 58 minutes in the technical program area.

"Our analysis of the data is ongoing, and we think we will find some interesting correlations that will help the SC conference as it plans for future conferences," said Donna Cox, head of NCSA's Experimental Technologies division. "For example, we will be able to see how well different program content held the attention of the participants by looking at their ebb and flow during the course of different talks and presentations."

"The challenge was to develop a system that would provide added value services to the conference participants and yet would be very non-intrusive and easy to use while tracking every move of each participant on the conference floor," said Volodymyr Kindratenko, IntelliBadge project technical lead. "Thus, it was decided that participation in the project would be voluntary and that the privacy and protection of all data provided by the participants would be our first priority."

Kindratenko added that for the vast majority of IntelliBadge participants privacy and legal issues were not major concerns. Just under 98 percent of all participants provided their names during the registration process and 95.5 percent of participants also allowed others to access their names and profiles. Some participants said they wanted to be tracked with more precision.

"The participants had faith in the integrity of the sponsors—in IEEE, ACM, SC2002, and NCSA," said Cox. "We made it clear we would respect their privacy and that none of the information would be sold, and they trusted us. I'm not sure if that would be the case with other conferences."

The IntelliBadge project also proved to be a useful community building tool, according to Dan Reed, director of NCSA, technical program chair for SC2002, and the originator of the IntelliBadge project.

"One of the best features of IntelliBadge was that it helped people with common interests identify each other, get together in person, and have a chance to talk informally about their ideas," he said. "That's one of the main reasons to come to a conference like SC, and the role of IntelliBadge in this kind of community building was very important."

Like all experiments, the project faced some technical difficulties. The immense size of the Baltimore Convention Center and the irregular shape of some rooms sometimes made tracking difficult. Real-time visualizations sometimes pushed the IntelliBadge computers beyond their limits because so much happened at once during the course of the typical day. "Installing and networking the equipment was a significant challenge and we were very thankful to SCinet and to Savi for helping us," said Kindratenko. "Once we got the equipment working, our biggest problem was with overlapping readers resulting in duplicate records and more data than was expected. Although our software was generally robust enough to deal with this issue, our database PC sustained a heavy load during the entire conference."

IntelliBadge was powered by commodity PCs, rather than connected to any supercomputing system, according to Cox. The SC2002 experience shows that presenting real-time visualizations of the flow of people across time and space could benefit from high-performance computing systems, she added.

"Our experience suggests that uniting remote sensing with supercomputing and real-time visualization could mean more data and more interesting presentations of the data," Cox said. "What would be really interesting is combining the remote sensing technology with the most creative visualization ideas and the power of high-performance computing. This would give participants a truly unique look at conferences that they thought they knew everything about."

What next for IntelliBadge? First, the NCSA team will continue to analyze the data to find correlations between variables, such as interests, time spent in technical sessions, time spent on the exhibit floor, and the number of technical sessions attended. They will present their findings to the SC2002 and 2003 planning committees at a January meeting.

In the long run, technologies that combine remote sensing and databases hold great appeal for industry. Businesses would love to know more about the people who stop by their booths at trade shows, including the amount of time they spend there and the products they look at. IntelliBadge is the type of technology that could reveal such data.

"In the business sector, the data becomes a valuable commodity," said Reed. "At that point, there are a lot of privacy and security issues that need to be figured out, and the private sector as well as society as a whole will be dealing with these issues in the years to come. For SC2002, our idea was to enhance the conference experience and make it easier for people to come together, strengthen their communities, and build new communities. We succeeded in doing that."


Access Online | Posted 1-21-2003