'Knowledge Discovery' of New Materials

By Roland Piquepaille

Data mining is often used by businesses such as retailers to find isolated trends buried into the mountains of data gathered at their cash registers. But this process cannot really be applied to science for discovering new materials even if you have huge databases describing all your previous experiments. This is why an interdisciplinary team of researchers and engineers at Purdue University is working on 'knowledge discovery', a new computer-aided product design method that uses supercomputing, AI and large 3D displays. "Instead of mining for a nugget of gold, knowledge discovery is more like sifting through a warehouse filled with small gears, levers, etc., none of which is particularly valuable by itself. After appropriate assembly, however, a Rolex watch emerges from the disparate parts," said James Caruthers, a professor of chemical engineering. The system allows researchers to interact with their data in their own languages and uses a 12x7 feet tiled wall to display the results. Read more...

Here are selected excerpts of the Purdue University news release.

A team of researchers at Purdue led by Caruthers is developing a computer environment that allows experts to talk naturally in their specific scientific language. That way, the researchers don't have to deal with computerese and can take full advantage of the most advanced visualization capabilities to become more engaged in the scientific discovery process, Caruthers said.

Such a system could become crucial for enabling scientists to deal with the recent explosion of data now available to them. The source of this flood of data is "high-throughput" experimentation, in which hundreds or thousands of experiments are conducted simultaneously in tiny vessels that are sometimes as small as a few human hairs. Having so much information presents a challenge: it is difficult for researchers to find what they are looking for within this huge sea of data.

And even if they have lots of data, what they really want is knowledge.

Purdue researchers believe they have a solution to the problem. They are developing a method to extract knowledge from data, promising to speed up the process of discovery in many areas of research, including work aimed at creating new drugs, fuel additives, catalysts and rubber compounds.

The method, called "discovery informatics," enables researchers to test new theories on the fly and literally see how well their concepts might work in real time via a three-dimensional display, said Venkat Venkatasubramanian, another professor of chemical engineering working to develop the new system.

This news release is pretty long, so let's just look here at the visualization part of the method.

Researchers in Purdue's e-Enterprise Center helped the chemical engineers create software prototypes needed to manage huge amounts of data and simulations, turning the information into interactive images, said Joseph Pekny, director of the e-Enterprise Center and a professor of chemical engineering.

Then information technology experts use supercomputers to run the complex software for applications such as predicting chemical reactions and then "visualizing" such data on a three-dimensional, 12-foot-wide, 7-foot-high display in the Envision Center, said Gary Bertoline, associate vice president for discovery resources at ITaP and a professor of computer graphics technology in Purdue's School of Technology.

"We are helping them look at large amounts of data all at the same time," said Laura Arns, a visualization and computer graphics application engineer at ITaP. "You can display information in stereo, in which the left and the right eye each get their own pictures, and you get a 3-D depth effect. To see the 3-D visualization, you wear special glasses that are like sun glasses."

Here is a photograph of a student reviewing and interacting with a chemical catalyst. (Credit: Purdue News Service; Photo: David Umberger) Please note that he doesn't even wear special glasses, so he's not working with stereo images.

The concept looks cool, but does it really lead to the creation of new products? Here is Caruthers's conclusion.

"Discovery requires human beings making intuitive leaps," Caruthers said. "You try one thing. It doesn't work, you try something else. Sometimes you go off in an entirely new direction. But this process is very inefficient. What we are doing is enhancing the efficiency of this process, assisting the intuitive human mind by providing massive data and computing power."

Source: Purdue University News, October 19, 2004